38 统计学习介绍

统计学习介绍的主要参考书: (James et al. 2013): Gareth James, Daniela Witten, Trevor Hastie, Robert Tibshirani(2013) An Introduction to Statistical Learning: with Applications in R, Springer.

调入需要的扩展包:

library(leaps) # 全子集回归
library(ISLR) # 参考书对应的包
library(glmnet) # 岭回归和lasso
## 载入需要的程辑包:Matrix
## Loaded glmnet 4.1-2
library(tree) # 树回归
library(randomForest) # 随机森林和装袋法
## randomForest 4.6-14
## Type rfNews() to see new features/changes/bug fixes.
library(MASS)
library(gbm) # boosting
## Loaded gbm 2.1.8
library(e1071) # svm

38.1 统计学习的基本概念和一般步骤

38.1.1 统计学习的基本概念和方法

统计学习(statistical learning), 也有数据挖掘(data mining),机器学习(machine learning)等称呼。 主要目的是用一些计算机算法从大量数据中发现知识。 方兴未艾的数据科学就以统计学习为重要支柱。 方法分为有监督(supervised)学习与无监督(unsupervised)学习。

无监督学习方法如聚类问题、购物篮问题、主成分分析等。

有监督学习即统计中回归分析和判别分析解决的问题, 现在又有树回归、树判别、随机森林、lasso、支持向量机、 神经网络、贝叶斯网络、排序算法等许多方法。

无监督学习在给了数据之后, 直接从数据中发现规律, 比如聚类分析是发现数据中的聚集和分组现象, 购物篮分析是从数据中找到更多的共同出现的条目 (比如购买啤酒的用户也有较大可能购买火腿肠)。

有监督学习方法众多。 通常,需要把数据分为训练样本和检验样本, 训练样本的因变量(数值型或分类型)是已知的, 根据训练样本中自变量和因变量的关系训练出一个回归函数, 此函数以自变量为输入, 可以输出因变量的预测值。

训练出的函数有可能是有简单表达式的(例如,logistic回归)、 有参数众多的表达式的(如神经网络), 也有可能是依赖于所有训练样本而无法写出表达式的(例如k近邻分类)。

38.1.2 偏差与方差折衷

对回归问题,经常使用均方误差\(E|Ey - \hat y|^2\)来衡量精度。 对分类问题,经常使用分类准确率等来衡量精度。 易见\(E|Ey - \hat y|^2 = \text{Var}(\hat y) + (E\hat y - E y)^2\),所以均方误差可以分解为 \[ \text{均方误差} = \text{方差} + \text{偏差}^2, \]

训练的回归函数如果仅考虑对训练样本解释尽可能好, 就会使得估计结果方差很大,在对检验样本进行计算时因方差大而导致很大的误差, 所以选取的回归函数应该尽可能简单。

如果选取的回归函数过于简单而实际上自变量与因变量关系比较复杂, 就会使得估计的回归函数偏差比较大, 这样在对检验样本进行计算时也会有比较大的误差。

所以,在有监督学习时, 回归函数的复杂程度是一个很关键的量, 太复杂和太简单都可能导致差的结果, 需要找到一个折衷的值。

复杂程度在线性回归中就是自变量个数, 在一元曲线拟合中就是曲线的不光滑程度。 在其它指标类似的情况下,简单的模型更稳定、可解释更好, 所以统计学特别重视模型的简化。

38.1.3 交叉验证

即使是在从训练样本中修炼(估计)回归函数时, 也需要适当地选择模型的复杂度。 仅考虑对训练数据的拟合程度是不够的, 这会造成过度拟合问题。

为了相对客观地度量模型的预报误差, 假设训练样本有\(n\)个观测, 可以留出第一个观测不用, 用剩余的\(n-1\)个观测建模,然后预测第一个观测的因变量值, 得到一个误差;对每个观测都这样做, 就可以得到\(n\)个误差。 这样的方法叫做留一法。

更常用的是五折或十折交叉验证。 假设训练集有\(n\)个观测, 将其均分成\(10\)分, 保留第一份不用, 将其余九份合并在一起用来建模,然后预报第一份; 对每一份都这样做, 也可以得到\(n\)个误差, 这叫做十折(ten-fold)交叉验证方法。

因为要预报的数据没有用来建模, 交叉验证得到的误差估计更准确。

38.1.4 一般步骤

一个有监督的统计学习项目, 大致上按如下步骤进行:

  • 将数据拆分为训练集和测试集;
  • 在训练集上比较不同的模型, 可以用交叉验证方法比较;
  • 使用训练集上找到的最优模型, 对测试集进行测试;
  • 在整个数据集上使用找到的最优模型, 但预测效果的指标使用测试集上得到的预测效果指标。

38.2 Hitters数据分析

考虑ISLR包的Hitters数据集。 此数据集有322个运动员的20个变量的数据, 其中的变量Salary(工资)是我们关心的。 变量包括:

names(Hitters)
##  [1] "AtBat"     "Hits"      "HmRun"     "Runs"      "RBI"       "Walks"     "Years"     "CAtBat"    "CHits"     "CHmRun"    "CRuns"     "CRBI"      "CWalks"    "League"    "Division"  "PutOuts"   "Assists"   "Errors"    "Salary"    "NewLeague"

数据集的详细变量信息如下:

str(Hitters)
## 'data.frame':    322 obs. of  20 variables:
##  $ AtBat    : int  293 315 479 496 321 594 185 298 323 401 ...
##  $ Hits     : int  66 81 130 141 87 169 37 73 81 92 ...
##  $ HmRun    : int  1 7 18 20 10 4 1 0 6 17 ...
##  $ Runs     : int  30 24 66 65 39 74 23 24 26 49 ...
##  $ RBI      : int  29 38 72 78 42 51 8 24 32 66 ...
##  $ Walks    : int  14 39 76 37 30 35 21 7 8 65 ...
##  $ Years    : int  1 14 3 11 2 11 2 3 2 13 ...
##  $ CAtBat   : int  293 3449 1624 5628 396 4408 214 509 341 5206 ...
##  $ CHits    : int  66 835 457 1575 101 1133 42 108 86 1332 ...
##  $ CHmRun   : int  1 69 63 225 12 19 1 0 6 253 ...
##  $ CRuns    : int  30 321 224 828 48 501 30 41 32 784 ...
##  $ CRBI     : int  29 414 266 838 46 336 9 37 34 890 ...
##  $ CWalks   : int  14 375 263 354 33 194 24 12 8 866 ...
##  $ League   : Factor w/ 2 levels "A","N": 1 2 1 2 2 1 2 1 2 1 ...
##  $ Division : Factor w/ 2 levels "E","W": 1 2 2 1 1 2 1 2 2 1 ...
##  $ PutOuts  : int  446 632 880 200 805 282 76 121 143 0 ...
##  $ Assists  : int  33 43 82 11 40 421 127 283 290 0 ...
##  $ Errors   : int  20 10 14 3 4 25 7 9 19 0 ...
##  $ Salary   : num  NA 475 480 500 91.5 750 70 100 75 1100 ...
##  $ NewLeague: Factor w/ 2 levels "A","N": 1 2 1 2 2 1 1 1 2 1 ...

希望以Salary为因变量,查看其缺失值个数:

sum( is.na(Hitters$Salary) )
## [1] 59

为简单起见,去掉有缺失值的观测:

d <- na.omit(Hitters); dim(d)
## [1] 263  20

38.2.1 回归自变量选择

38.2.1.1 最优子集选择

用leaps包的regsubsets()函数计算最优子集回归, 办法是对某个试验性的子集自变量个数\(\hat p\)值, 都找到\(\hat p\)固定情况下残差平方和最小的变量子集, 这样只要在这些不同\(\hat p\)的最优子集中挑选就可以了。 挑选可以用AIC、BIC等方法。

可以先进行一个包含所有自变量的全集回归:

regfit.full <- regsubsets(Salary ~ ., data=d, nvmax=19)
reg.summary <- summary(regfit.full)
reg.summary
## Subset selection object
## Call: regsubsets.formula(Salary ~ ., data = d, nvmax = 19)
## 19 Variables  (and intercept)
##            Forced in Forced out
## AtBat          FALSE      FALSE
## Hits           FALSE      FALSE
## HmRun          FALSE      FALSE
## Runs           FALSE      FALSE
## RBI            FALSE      FALSE
## Walks          FALSE      FALSE
## Years          FALSE      FALSE
## CAtBat         FALSE      FALSE
## CHits          FALSE      FALSE
## CHmRun         FALSE      FALSE
## CRuns          FALSE      FALSE
## CRBI           FALSE      FALSE
## CWalks         FALSE      FALSE
## LeagueN        FALSE      FALSE
## DivisionW      FALSE      FALSE
## PutOuts        FALSE      FALSE
## Assists        FALSE      FALSE
## Errors         FALSE      FALSE
## NewLeagueN     FALSE      FALSE
## 1 subsets of each size up to 19
## Selection Algorithm: exhaustive
##           AtBat Hits HmRun Runs RBI Walks Years CAtBat CHits CHmRun CRuns CRBI CWalks LeagueN DivisionW PutOuts Assists Errors NewLeagueN
## 1  ( 1 )  " "   " "  " "   " "  " " " "   " "   " "    " "   " "    " "   "*"  " "    " "     " "       " "     " "     " "    " "       
## 2  ( 1 )  " "   "*"  " "   " "  " " " "   " "   " "    " "   " "    " "   "*"  " "    " "     " "       " "     " "     " "    " "       
## 3  ( 1 )  " "   "*"  " "   " "  " " " "   " "   " "    " "   " "    " "   "*"  " "    " "     " "       "*"     " "     " "    " "       
## 4  ( 1 )  " "   "*"  " "   " "  " " " "   " "   " "    " "   " "    " "   "*"  " "    " "     "*"       "*"     " "     " "    " "       
## 5  ( 1 )  "*"   "*"  " "   " "  " " " "   " "   " "    " "   " "    " "   "*"  " "    " "     "*"       "*"     " "     " "    " "       
## 6  ( 1 )  "*"   "*"  " "   " "  " " "*"   " "   " "    " "   " "    " "   "*"  " "    " "     "*"       "*"     " "     " "    " "       
## 7  ( 1 )  " "   "*"  " "   " "  " " "*"   " "   "*"    "*"   "*"    " "   " "  " "    " "     "*"       "*"     " "     " "    " "       
## 8  ( 1 )  "*"   "*"  " "   " "  " " "*"   " "   " "    " "   "*"    "*"   " "  "*"    " "     "*"       "*"     " "     " "    " "       
## 9  ( 1 )  "*"   "*"  " "   " "  " " "*"   " "   "*"    " "   " "    "*"   "*"  "*"    " "     "*"       "*"     " "     " "    " "       
## 10  ( 1 ) "*"   "*"  " "   " "  " " "*"   " "   "*"    " "   " "    "*"   "*"  "*"    " "     "*"       "*"     "*"     " "    " "       
## 11  ( 1 ) "*"   "*"  " "   " "  " " "*"   " "   "*"    " "   " "    "*"   "*"  "*"    "*"     "*"       "*"     "*"     " "    " "       
## 12  ( 1 ) "*"   "*"  " "   "*"  " " "*"   " "   "*"    " "   " "    "*"   "*"  "*"    "*"     "*"       "*"     "*"     " "    " "       
## 13  ( 1 ) "*"   "*"  " "   "*"  " " "*"   " "   "*"    " "   " "    "*"   "*"  "*"    "*"     "*"       "*"     "*"     "*"    " "       
## 14  ( 1 ) "*"   "*"  "*"   "*"  " " "*"   " "   "*"    " "   " "    "*"   "*"  "*"    "*"     "*"       "*"     "*"     "*"    " "       
## 15  ( 1 ) "*"   "*"  "*"   "*"  " " "*"   " "   "*"    "*"   " "    "*"   "*"  "*"    "*"     "*"       "*"     "*"     "*"    " "       
## 16  ( 1 ) "*"   "*"  "*"   "*"  "*" "*"   " "   "*"    "*"   " "    "*"   "*"  "*"    "*"     "*"       "*"     "*"     "*"    " "       
## 17  ( 1 ) "*"   "*"  "*"   "*"  "*" "*"   " "   "*"    "*"   " "    "*"   "*"  "*"    "*"     "*"       "*"     "*"     "*"    "*"       
## 18  ( 1 ) "*"   "*"  "*"   "*"  "*" "*"   "*"   "*"    "*"   " "    "*"   "*"  "*"    "*"     "*"       "*"     "*"     "*"    "*"       
## 19  ( 1 ) "*"   "*"  "*"   "*"  "*" "*"   "*"   "*"    "*"   "*"    "*"   "*"  "*"    "*"     "*"       "*"     "*"     "*"    "*"

这里用nvmax=指定了允许所有的自变量都参加, 缺省行为是限制最多个数的。 上述结果表格中每一行给出了固定\(\hat p\)条件下的最优子集。

试比较这些最优模型的BIC值:

reg.summary$bic
##  [1]  -90.84637 -128.92622 -135.62693 -141.80892 -144.07143 -147.91690 -145.25594 -147.61525 -145.44316 -143.21651 -138.86077 -133.87283 -128.77759 -123.64420 -118.21832 -112.81768 -107.35339 -101.86391  -96.30412
plot(reg.summary$bic)
Hitters数据最优子集回归BIC

图38.1: Hitters数据最优子集回归BIC

其中\(\hat p=6, 8\)的值相近,都很低, 取\(\hat p=6\)。 用coef()id=6指定第六种子集:

coef(regfit.full, id=6)
##  (Intercept)        AtBat         Hits        Walks         CRBI    DivisionW      PutOuts 
##   91.5117981   -1.8685892    7.6043976    3.6976468    0.6430169 -122.9515338    0.2643076

这种方法实现了选取BIC最小的自变量子集。

38.2.1.2 逐步回归方法

在用做了全集回归后, 把全集回归结果输入到函数中可以执行逐步回归。 如:

lm.full <- lm(Salary ~ ., data=d)
print(summary(lm.full))
## 
## Call:
## lm(formula = Salary ~ ., data = d)
## 
## Residuals:
##     Min      1Q  Median      3Q     Max 
## -907.62 -178.35  -31.11  139.09 1877.04 
## 
## Coefficients:
##               Estimate Std. Error t value Pr(>|t|)    
## (Intercept)  163.10359   90.77854   1.797 0.073622 .  
## AtBat         -1.97987    0.63398  -3.123 0.002008 ** 
## Hits           7.50077    2.37753   3.155 0.001808 ** 
## HmRun          4.33088    6.20145   0.698 0.485616    
## Runs          -2.37621    2.98076  -0.797 0.426122    
## RBI           -1.04496    2.60088  -0.402 0.688204    
## Walks          6.23129    1.82850   3.408 0.000766 ***
## Years         -3.48905   12.41219  -0.281 0.778874    
## CAtBat        -0.17134    0.13524  -1.267 0.206380    
## CHits          0.13399    0.67455   0.199 0.842713    
## CHmRun        -0.17286    1.61724  -0.107 0.914967    
## CRuns          1.45430    0.75046   1.938 0.053795 .  
## CRBI           0.80771    0.69262   1.166 0.244691    
## CWalks        -0.81157    0.32808  -2.474 0.014057 *  
## LeagueN       62.59942   79.26140   0.790 0.430424    
## DivisionW   -116.84925   40.36695  -2.895 0.004141 ** 
## PutOuts        0.28189    0.07744   3.640 0.000333 ***
## Assists        0.37107    0.22120   1.678 0.094723 .  
## Errors        -3.36076    4.39163  -0.765 0.444857    
## NewLeagueN   -24.76233   79.00263  -0.313 0.754218    
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
## 
## Residual standard error: 315.6 on 243 degrees of freedom
## Multiple R-squared:  0.5461, Adjusted R-squared:  0.5106 
## F-statistic: 15.39 on 19 and 243 DF,  p-value: < 2.2e-16
lm.step <- step(lm.full)
## Start:  AIC=3046.02
## Salary ~ AtBat + Hits + HmRun + Runs + RBI + Walks + Years + 
##     CAtBat + CHits + CHmRun + CRuns + CRBI + CWalks + League + 
##     Division + PutOuts + Assists + Errors + NewLeague
## 
##             Df Sum of Sq      RSS    AIC
## - CHmRun     1      1138 24201837 3044.0
## - CHits      1      3930 24204629 3044.1
## - Years      1      7869 24208569 3044.1
## - NewLeague  1      9784 24210484 3044.1
## - RBI        1     16076 24216776 3044.2
## - HmRun      1     48572 24249272 3044.6
## - Errors     1     58324 24259023 3044.7
## - League     1     62121 24262821 3044.7
## - Runs       1     63291 24263990 3044.7
## - CRBI       1    135439 24336138 3045.5
## - CAtBat     1    159864 24360564 3045.8
## <none>                   24200700 3046.0
## - Assists    1    280263 24480963 3047.1
## - CRuns      1    374007 24574707 3048.1
## - CWalks     1    609408 24810108 3050.6
## - Division   1    834491 25035190 3052.9
## - AtBat      1    971288 25171987 3054.4
## - Hits       1    991242 25191941 3054.6
## - Walks      1   1156606 25357305 3056.3
## - PutOuts    1   1319628 25520328 3058.0
## 
## Step:  AIC=3044.03
## Salary ~ AtBat + Hits + HmRun + Runs + RBI + Walks + Years + 
##     CAtBat + CHits + CRuns + CRBI + CWalks + League + Division + 
##     PutOuts + Assists + Errors + NewLeague
## 
##             Df Sum of Sq      RSS    AIC
## - Years      1      7609 24209447 3042.1
## - NewLeague  1     10268 24212106 3042.2
## - CHits      1     14003 24215840 3042.2
## - RBI        1     14955 24216793 3042.2
## - HmRun      1     52777 24254614 3042.6
## - Errors     1     59530 24261367 3042.7
## - League     1     63407 24265244 3042.7
## - Runs       1     64860 24266698 3042.7
## - CAtBat     1    174992 24376830 3043.9
## <none>                   24201837 3044.0
## - Assists    1    285766 24487603 3045.1
## - CRuns      1    611358 24813196 3048.6
## - CWalks     1    645627 24847464 3049.0
## - Division   1    834637 25036474 3050.9
## - CRBI       1    864220 25066057 3051.3
## - AtBat      1    970861 25172699 3052.4
## - Hits       1   1025981 25227819 3052.9
## - Walks      1   1167378 25369216 3054.4
## - PutOuts    1   1325273 25527110 3056.1
## 
## Step:  AIC=3042.12
## Salary ~ AtBat + Hits + HmRun + Runs + RBI + Walks + CAtBat + 
##     CHits + CRuns + CRBI + CWalks + League + Division + PutOuts + 
##     Assists + Errors + NewLeague
## 
##             Df Sum of Sq      RSS    AIC
## - NewLeague  1      9931 24219377 3040.2
## - RBI        1     15989 24225436 3040.3
## - CHits      1     18291 24227738 3040.3
## - HmRun      1     54144 24263591 3040.7
## - Errors     1     57312 24266759 3040.7
## - Runs       1     63172 24272619 3040.8
## - League     1     65732 24275178 3040.8
## <none>                   24209447 3042.1
## - CAtBat     1    266205 24475652 3043.0
## - Assists    1    293479 24502926 3043.3
## - CRuns      1    646350 24855797 3047.1
## - CWalks     1    649269 24858716 3047.1
## - Division   1    827511 25036958 3049.0
## - CRBI       1    872121 25081568 3049.4
## - AtBat      1    968713 25178160 3050.4
## - Hits       1   1018379 25227825 3050.9
## - Walks      1   1164536 25373983 3052.5
## - PutOuts    1   1334525 25543972 3054.2
## 
## Step:  AIC=3040.22
## Salary ~ AtBat + Hits + HmRun + Runs + RBI + Walks + CAtBat + 
##     CHits + CRuns + CRBI + CWalks + League + Division + PutOuts + 
##     Assists + Errors
## 
##            Df Sum of Sq      RSS    AIC
## - RBI       1     15800 24235177 3038.4
## - CHits     1     15859 24235237 3038.4
## - Errors    1     54505 24273883 3038.8
## - HmRun     1     54938 24274316 3038.8
## - Runs      1     62294 24281671 3038.9
## - League    1    107479 24326856 3039.4
## <none>                  24219377 3040.2
## - CAtBat    1    261336 24480713 3041.1
## - Assists   1    295536 24514914 3041.4
## - CWalks    1    648860 24868237 3045.2
## - CRuns     1    661449 24880826 3045.3
## - Division  1    824672 25044049 3047.0
## - CRBI      1    880429 25099806 3047.6
## - AtBat     1    999057 25218434 3048.9
## - Hits      1   1034463 25253840 3049.2
## - Walks     1   1157205 25376583 3050.5
## - PutOuts   1   1335173 25554550 3052.3
## 
## Step:  AIC=3038.4
## Salary ~ AtBat + Hits + HmRun + Runs + Walks + CAtBat + CHits + 
##     CRuns + CRBI + CWalks + League + Division + PutOuts + Assists + 
##     Errors
## 
##            Df Sum of Sq      RSS    AIC
## - CHits     1     13483 24248660 3036.5
## - HmRun     1     44586 24279763 3036.9
## - Runs      1     54057 24289234 3037.0
## - Errors    1     57656 24292833 3037.0
## - League    1    108644 24343821 3037.6
## <none>                  24235177 3038.4
## - CAtBat    1    252756 24487934 3039.1
## - Assists   1    294674 24529851 3039.6
## - CWalks    1    639690 24874868 3043.2
## - CRuns     1    693535 24928712 3043.8
## - Division  1    808984 25044161 3045.0
## - CRBI      1    893830 25129008 3045.9
## - Hits      1   1034884 25270061 3047.4
## - AtBat     1   1042798 25277975 3047.5
## - Walks     1   1145013 25380191 3048.5
## - PutOuts   1   1340713 25575890 3050.6
## 
## Step:  AIC=3036.54
## Salary ~ AtBat + Hits + HmRun + Runs + Walks + CAtBat + CRuns + 
##     CRBI + CWalks + League + Division + PutOuts + Assists + Errors
## 
##            Df Sum of Sq      RSS    AIC
## - HmRun     1     40487 24289148 3035.0
## - Errors    1     51930 24300590 3035.1
## - Runs      1     79343 24328003 3035.4
## - League    1    114742 24363402 3035.8
## <none>                  24248660 3036.5
## - Assists   1    283442 24532103 3037.6
## - CAtBat    1    613356 24862016 3041.1
## - Division  1    801474 25050134 3043.1
## - CRBI      1    903248 25151908 3044.2
## - CWalks    1   1011953 25260613 3045.3
## - Walks     1   1246164 25494824 3047.7
## - AtBat     1   1339620 25588280 3048.7
## - CRuns     1   1390808 25639469 3049.2
## - PutOuts   1   1406023 25654684 3049.4
## - Hits      1   1607990 25856650 3051.4
## 
## Step:  AIC=3034.98
## Salary ~ AtBat + Hits + Runs + Walks + CAtBat + CRuns + CRBI + 
##     CWalks + League + Division + PutOuts + Assists + Errors
## 
##            Df Sum of Sq      RSS    AIC
## - Errors    1     44085 24333232 3033.5
## - Runs      1     49068 24338215 3033.5
## - League    1    103837 24392985 3034.1
## <none>                  24289148 3035.0
## - Assists   1    247002 24536150 3035.6
## - CAtBat    1    652746 24941894 3040.0
## - Division  1    795643 25084791 3041.5
## - CWalks    1    982896 25272044 3043.4
## - Walks     1   1205823 25494971 3045.7
## - AtBat     1   1300972 25590120 3046.7
## - CRuns     1   1351200 25640348 3047.2
## - CRBI      1   1353507 25642655 3047.2
## - PutOuts   1   1429006 25718154 3048.0
## - Hits      1   1574140 25863288 3049.5
## 
## Step:  AIC=3033.46
## Salary ~ AtBat + Hits + Runs + Walks + CAtBat + CRuns + CRBI + 
##     CWalks + League + Division + PutOuts + Assists
## 
##            Df Sum of Sq      RSS    AIC
## - Runs      1     54113 24387345 3032.0
## - League    1     91269 24424501 3032.4
## <none>                  24333232 3033.5
## - Assists   1    220010 24553242 3033.8
## - CAtBat    1    650513 24983746 3038.4
## - Division  1    799455 25132687 3040.0
## - CWalks    1    971260 25304493 3041.8
## - Walks     1   1239533 25572765 3044.5
## - CRBI      1   1331672 25664904 3045.5
## - CRuns     1   1361070 25694302 3045.8
## - AtBat     1   1378592 25711824 3045.9
## - PutOuts   1   1391660 25724892 3046.1
## - Hits      1   1649291 25982523 3048.7
## 
## Step:  AIC=3032.04
## Salary ~ AtBat + Hits + Walks + CAtBat + CRuns + CRBI + CWalks + 
##     League + Division + PutOuts + Assists
## 
##            Df Sum of Sq      RSS    AIC
## - League    1    113056 24500402 3031.3
## <none>                  24387345 3032.0
## - Assists   1    280689 24668034 3033.1
## - CAtBat    1    596622 24983967 3036.4
## - Division  1    780369 25167714 3038.3
## - CWalks    1    946687 25334032 3040.1
## - Walks     1   1212997 25600342 3042.8
## - CRuns     1   1334397 25721742 3044.1
## - CRBI      1   1361339 25748684 3044.3
## - PutOuts   1   1455210 25842555 3045.3
## - AtBat     1   1522760 25910105 3046.0
## - Hits      1   1718870 26106215 3047.9
## 
## Step:  AIC=3031.26
## Salary ~ AtBat + Hits + Walks + CAtBat + CRuns + CRBI + CWalks + 
##     Division + PutOuts + Assists
## 
##            Df Sum of Sq      RSS    AIC
## <none>                  24500402 3031.3
## - Assists   1    313650 24814051 3032.6
## - CAtBat    1    534156 25034558 3034.9
## - Division  1    798473 25298875 3037.7
## - CWalks    1    965875 25466276 3039.4
## - CRuns     1   1265082 25765484 3042.5
## - Walks     1   1290168 25790569 3042.8
## - CRBI      1   1326770 25827172 3043.1
## - PutOuts   1   1551523 26051925 3045.4
## - AtBat     1   1589780 26090181 3045.8
## - Hits      1   1716068 26216469 3047.1
print(lm.step)
## 
## Call:
## lm(formula = Salary ~ AtBat + Hits + Walks + CAtBat + CRuns + 
##     CRBI + CWalks + Division + PutOuts + Assists, data = d)
## 
## Coefficients:
## (Intercept)        AtBat         Hits        Walks       CAtBat        CRuns         CRBI       CWalks    DivisionW      PutOuts      Assists  
##    162.5354      -2.1687       6.9180       5.7732      -0.1301       1.4082       0.7743      -0.8308    -112.3801       0.2974       0.2832

最后保留了10个自变量。

38.2.1.3 划分训练集与测试集

在整个数据集中随机选取一部分作为训练集,其余作为测试集。 下面的程序把原始数据一分为二:

set.seed(1)
train <- sample(nrow(d), size = round(nrow(d)/2))
test <- -train

仅用训练集估计模型。 为了在测试集上用模型进行预报并估计预测均方误差, 需要自己写一个预测函数:

predict.regsubsets <- function(object, newdata, id, ...){
  form <- as.formula(object$call[[2]])
  mat <- model.matrix(form, newdata)
  coefi <- coef(object, id=id)
  xvars <- names(coefi)
  mat[, xvars] %*% coefi
}

然后,对每个子集大小,用最优子集在测试集上进行预报, 计算均方误差:

regfit.best <- regsubsets( Salary ~ ., data=d[train,], nvmax=19 )
val.errors <- rep(as.numeric(NA), 19)
for(i in 1:19){
  #pred <- predict.regsubsets(regfit.best, newdata=d[test,], id=i)
  pred <- predict(regfit.best, newdata=d[test,], id=i)
  val.errors[i] <- mean( (d[test, 'Salary'] - pred)^2 )
}
print(val.errors)
##  [1] 188190.9 163306.2 152365.0 164857.0 152100.7 147120.0 148833.0 155546.5 167429.2 169949.1 173607.9 173039.5 168450.4 169300.5 169139.3 173575.1 175216.2 175080.2 175057.5
best.id <- which.min(val.errors); best.id
## [1] 6

用测试集得到的最优子集大小为6。 模型子集和回归系数为:

coef(regfit.best, id=best.id)
##  (Intercept)        Walks       CAtBat        CHits       CHmRun    DivisionW      PutOuts 
##  179.4442609    4.1205817   -0.5508342    2.1670021    2.3479409 -126.3067258    0.1840943

38.2.1.4 用10折交叉验证方法选择最优子集

下列程序对数据中每一行分配一个折号:

k <- 10
set.seed(1)
folds <- sample(1:k, nrow(d), replace=TRUE)

下面,对10折中每一折都分别当作测试集一次, 得到不同子集大小的均方误差:

cv.errors <- matrix( as.numeric(NA), k, 19, dimnames=list(NULL, paste(1:19)) )
for(j in 1:k){ # 对
  best.fit <- regsubsets(Salary ~ ., data=d[folds != j,], nvmax=19)
  for(i in 1:19){
    pred <- predict( best.fit, d[folds==j,], id=i)
    cv.errors[j, i] <- mean( (d[folds==j, 'Salary'] - pred)^2 )
  }
}
head(cv.errors)
##              1         2         3         4         5         6         7         8         9        10        11        12        13        14        15        16        17        18        19
## [1,]  98623.24 115600.61 120884.31 113831.63 120728.51 122922.93 155507.25 137753.36 149198.01 153332.89 155702.91 155842.88 158755.87 156037.17 157739.46 155548.96 156688.01 156860.92 156976.98
## [2,] 155320.11 100425.87 168838.35 159729.47 145895.71 123555.25 119983.35  96609.16  99057.32  80375.78  91290.74  92292.69 100498.84 101562.45 104621.38 100922.27 102198.69 105318.26 106064.89
## [3,] 124151.77  68833.50  69392.29  77221.37  83802.82  70125.41  68997.77  64143.70  65813.14  65120.27  68160.94  70263.77  69765.81  68987.54  69471.32  69294.21  69199.91  68866.84  69195.74
## [4,] 232191.41 279001.29 294568.10 288765.81 276972.83 260121.22 276413.09 259923.88 270151.18 263492.31 259154.53 269017.80 265468.90 269666.65 265518.87 267240.44 267771.74 267670.66 267717.80
## [5,] 115397.35  96807.44 108421.66 104933.55  99561.69  86103.05  89345.61  87693.15  91631.88  88763.37  89801.07  91070.44  92429.43  92821.15  95849.81  96513.70  95209.20  94952.21  94951.70
## [6,] 103839.30  75652.50  69962.31  58291.91  65893.45  64215.56  65800.88  61413.45  60200.70  59599.54  59831.90  60081.48  59662.51  60618.91  62540.03  62776.81  62717.77  62354.97  62268.97

cv.errors是一个\(10\times 19\)矩阵, 每行对应一折作为测试集的情形, 每列是一个子集大小, 元素值是测试均方误差。

对每列的10个元素求平均, 可以得到每个子集大小的平均均方误差:

mean.cv.errors <- apply(cv.errors, 2, mean)
mean.cv.errors
##        1        2        3        4        5        6        7        8        9       10       11       12       13       14       15       16       17       18       19 
## 149821.1 130922.0 139127.0 131028.8 131050.2 119538.6 124286.1 113580.0 115556.5 112216.7 113251.2 115755.9 117820.8 119481.2 120121.6 120074.3 120084.8 120085.8 120403.5
best.id <- which.min(mean.cv.errors)
plot(mean.cv.errors, type='b')
Hitters数据CV均方误差

图38.2: Hitters数据CV均方误差

这样找到的最优子集大小是10。 用这种方法找到最优子集大小后, 可以对全数据集重新建模但是选择最优子集大小为10:

reg.best <- regsubsets(Salary ~ ., data=d, nvmax=19)
coef(reg.best, id=best.id)
##  (Intercept)        AtBat         Hits        Walks       CAtBat        CRuns         CRBI       CWalks    DivisionW      PutOuts      Assists 
##  162.5354420   -2.1686501    6.9180175    5.7732246   -0.1300798    1.4082490    0.7743122   -0.8308264 -112.3800575    0.2973726    0.2831680

事实上, 划分训练集和验证集与交叉验证方法经常联合运用。 取一个固定的较小规模的测试集, 此测试集不用来作子集选择, 对训练集用交叉验证方法选择最优子集, 然后再测试集上验证。

38.2.2 岭回归

当自变量个数太多时,模型复杂度高, 可能有过度拟合, 模型不稳定。

一种方法是对较大的模型系数施加二次惩罚, 把最小二乘问题变成带有二次惩罚项的惩罚最小二乘问题: \[\begin{aligned} \min\; \sum_{i=1}^n \left( y_i - \beta_0 - \beta_1 x_{i1} - \dots - \beta_p x_{ip} \right)^2 + \lambda \sum_{j=1}^p \beta_j^2 . \end{aligned}\] 这比通常最小二乘得到的回归系数绝对值变小, 但是求解的稳定性增加了,避免了共线问题。

实际上, 与线性模型\(\boldsymbol Y = \boldsymbol X \boldsymbol\beta + \boldsymbol\varepsilon\) 的普通最小二乘解 \(\hat{\boldsymbol\beta} = (\boldsymbol X^T \boldsymbol X)^{-1} \boldsymbol X^T \boldsymbol Y\) 相比, 岭回归问题的解为 \[ \tilde{\boldsymbol\beta} = (\boldsymbol X^T \boldsymbol X + s \boldsymbol I)^{-1} \boldsymbol X^T \boldsymbol Y \] 其中\(\boldsymbol I\)为单位阵,\(s>0\)\(\lambda\)有关。

\(\lambda\)称为调节参数,\(\lambda\)越大,相当于模型复杂度越低。 适当选择\(\lambda\)可以在方差与偏差之间找到适当的折衷, 从而减小预测误差。

由于量纲问题,在不同自变量不可比时,数据集应该进行标准化。

用R的glmnet包计算岭回归。 用glmnet()函数, 指定参数alpha=0时执行的是岭回归。 用参数lambda=指定一个调节参数网格, 岭回归将在这些调节参数上计算。 用coef()从回归结果中取得不同调节参数对应的回归系数估计, 结果是一个矩阵,每列对应于一个调节参数。

仍采用上面去掉了缺失值的Hitters数据集结果d。

如下程序把回归的设计阵与因变量提取出来:

x <- model.matrix(Salary ~ ., d)[,-1]
y <- d$Salary

岭回归涉及到调节参数\(\lambda\)的选择, 为了绘图, 先选择\(\lambda\)的一个网格:

grid <- 10^seq(10, -2, length=100)

用所有数据针对这样的调节参数网格计算岭回归结果, 注意glmnet()函数允许调节参数\(\lambda\)输入多个值:

ridge.mod <- glmnet(x, y, alpha=0, lambda=grid)
dim(coef(ridge.mod))
## [1]  20 100

glmnet()函数默认对数据进行标准化。
coef()的结果是一个矩阵,每列对应一个调节参数值。

38.2.2.1 划分训练集与测试集

如下程序把数据分为一半训练、一半测试:

set.seed(1)
train <- sample(nrow(x), size = nrow(x)/2)
test <- (-train)
y.test <- y[test]

仅用测试集建立岭回归:

ridge.mod <- glmnet(x[train,], y[train], alpha=0, lambda=grid, thresh=1E-12)

用建立的模型对测试集进行预测,并计算调节参数等于4时的均方误差:

ridge.pred <- predict( ridge.mod, s=4, newx=x[test,] )
mean( (ridge.pred - y.test)^2 )
## [1] 142199.2

如果用因变量平均值作预测, 这是最差的预测:

mean( (mean(y[train]) - y.test)^2 )
## [1] 224669.9

\(\lambda=4\)的结果要好得多。 事实上,取\(\lambda\)接近正无穷时模型就相当于用因变量平均值预测。 取\(\lambda=0\)就相当于普通最小二乘回归(但是glmnet()是对输入数据要做标准化的)。

38.2.2.2 用10折交叉验证选取调节参数

仍使用训练集, 但训练集再进行交叉验证。 cv.glmnet()函数可以执行交叉验证。

set.seed(1)
cv.out <- cv.glmnet(x[train,], y[train], alpha=0)
plot(cv.out)
Hitters数据岭回归参数选择

图38.3: Hitters数据岭回归参数选择

bestlam <- cv.out$lambda.min

这样获得了最优调节参数\(\lambda=\) 326.0827885。 用最优调节参数对测试集作预测, 得到预测均方误差:

ridge.pred <- predict(ridge.mod, s=bestlam, newx=x[test,])
mean( (ridge.pred - y.test)^2 )
## [1] 139856.6

结果比\(\lambda=4\)略有改进。

最后,用选取的最优调节系数对全数据集建模, 得到相应的岭回归系数估计:

out <- glmnet(x, y, alpha=0)
predict(out, type='coefficients', s=bestlam)[1:20,]
##  (Intercept)        AtBat         Hits        HmRun         Runs          RBI        Walks        Years       CAtBat        CHits       CHmRun        CRuns         CRBI       CWalks      LeagueN    DivisionW      PutOuts      Assists       Errors   NewLeagueN 
##  15.44383135   0.07715547   0.85911581   0.60103107   1.06369007   0.87936105   1.62444616   1.35254780   0.01134999   0.05746654   0.40680157   0.11456224   0.12116504   0.05299202  22.09143189 -79.04032637   0.16619903   0.02941950  -1.36092945   9.12487767

38.2.3 Lasso回归

另一种对回归系数的惩罚是\(L_1\)惩罚: \[\begin{align} \min\; \sum_{i=1}^n \left( y_i - \beta_0 - \beta_1 x_{i1} - \dots - \beta_p x_{ip} \right)^2 + \lambda \sum_{j=1}^p |\beta_j| . \tag{38.1} \end{align}\] 奇妙地是,适当选择调节参数\(\lambda\),可以使得部分回归系数变成零, 达到了即减小回归系数的绝对值又挑选重要变量子集的效果。

事实上,(38.1)等价于约束最小值问题 \[\begin{aligned} & \min\; \sum_{i=1}^n \left( y_i - \beta_0 - \beta_1 x_{i1} - \dots - \beta_p x_{ip} \right)^2 \quad \text{s.t.} \\ & \sum_{j=1}^p |\beta_j| \leq s \end{aligned}\] 其中\(s\)\(\lambda\)一一对应。 这样的约束区域是带有顶点的凸集, 而目标函数是二次函数, 最小值点经常在约束区域顶点达到, 这些顶点是某些坐标等于零的点。 见图38.4

knitr::include_graphics("figs/lasso-min.png")
Lasso约束优化问题图示

图38.4: Lasso约束优化问题图示

对于每个调节参数\(\lambda\), 都应该解出(38.1)的相应解, 记为\(\hat{\boldsymbol\beta}(\lambda)\)。 幸运的是, 不需要对每个\(\lambda\)去解最小值问题(38.1), 存在巧妙的算法使得问题的计算量与求解一次最小二乘相仿。

通常选取\(\lambda\)的格子点,计算相应的惩罚回归系数。 用交叉验证方法估计预测的均方误差。 选取使得交叉验证均方误差最小的调节参数(一般R函数中已经作为选项)。

用R的glmnet包计算lasso。 用glmnet()函数, 指定参数alpha=1时执行的是lasso。 用参数lambda=指定一个调节参数网格, lasso将输出这些调节参数对应的结果。 对回归结果使用plot()函数可以画出调节参数变化时系数估计的变化情况。

仍使用gmlnet包的glmnet()函数计算Lasso回归, 指定一个调节参数网格(沿用前面的网格):

lasso.mod <- glmnet(x[train,], y[train], alpha=1, lambda=grid)
plot(lasso.mod)
## Warning in regularize.values(x, y, ties, missing(ties), na.rm = na.rm): collapsing to unique 'x' values
Hitters数据lasso轨迹

图38.5: Hitters数据lasso轨迹

对lasso结果使用plot()函数可以绘制延调节参数网格变化的各回归系数估计,横坐标不是调节参数而是调节参数对应的系数绝对值和, 可以看出随着系数绝对值和增大,实际是调节参数变小, 更多地自变量进入模型。

38.2.3.1 用交叉验证估计调节参数

按照前面划分的训练集与测试集, 仅使用训练集数据做交叉验证估计最优调节参数:

set.seed(1)
cv.out <- cv.glmnet(x[train,], y[train], alpha=1)
plot(cv.out)

bestlam <- cv.out$lambda.min; bestlam
## [1] 9.286955

得到调节参数估计后,对测试集计算预测均方误差:

lasso.pred <- predict(lasso.mod, s=bestlam, newx=x[test,])
mean( (lasso.pred - y.test)^2 )
## [1] 143673.6

这个效果比岭回归效果略差。

为了充分利用数据, 使用前面获得的最优调节参数, 对全数据集建模:

out <- glmnet(x, y, alpha=1, lambda=grid)
lasso.coef <- predict(out, type='coefficients', s=bestlam)[1:20,]; lasso.coef
##   (Intercept)         AtBat          Hits         HmRun          Runs           RBI         Walks         Years        CAtBat         CHits        CHmRun         CRuns          CRBI        CWalks       LeagueN     DivisionW       PutOuts       Assists        Errors    NewLeagueN 
##    1.27479059   -0.05497143    2.18034583    0.00000000    0.00000000    0.00000000    2.29192406   -0.33806109    0.00000000    0.00000000    0.02825013    0.21628385    0.41712537    0.00000000   20.28615023 -116.16755870    0.23752385    0.00000000   -0.85629148    0.00000000
lasso.coef[lasso.coef != 0]
##   (Intercept)         AtBat          Hits         Walks         Years        CHmRun         CRuns          CRBI       LeagueN     DivisionW       PutOuts        Errors 
##    1.27479059   -0.05497143    2.18034583    2.29192406   -0.33806109    0.02825013    0.21628385    0.41712537   20.28615023 -116.16755870    0.23752385   -0.85629148

选择的自变量子集有11个自变量。

38.2.4 树回归的简单演示

决策树方法按不同自变量的不同值, 分层地把训练集分组。 每层使用一个变量, 所以这样的分组构成一个二叉树表示。 为了预测一个观测的类归属, 找到它所属的组, 用组的类归属或大多数观测的类归属进行预测。 这样的方法称为决策树(decision tree)。 决策树方法既可以用于判别问题, 也可以用于回归问题,称为回归树。

决策树的好处是容易解释, 在自变量为分类变量时没有额外困难。 但预测准确率可能比其它有监督学习方法差。

改进方法包括装袋法(bagging)、随机森林(random forests)、 提升法(boosting)。 这些改进方法都是把许多棵树合并在一起, 通常能改善准确率但是可解释性变差。

对Hitters数据,用Years和Hits作因变量预测log(Salaray)。

仅取Hitters数据集的Salary, Years, Hits三个变量, 并仅保留完全观测:

d <- na.omit(Hitters[,c('Salary', 'Years', 'Hits')])
print(str(d))
## 'data.frame':    263 obs. of  3 variables:
##  $ Salary: num  475 480 500 91.5 750 ...
##  $ Years : int  14 3 11 2 11 2 3 2 13 10 ...
##  $ Hits  : int  81 130 141 87 169 37 73 81 92 159 ...
##  - attr(*, "na.action")= 'omit' Named int [1:59] 1 16 19 23 31 33 37 39 40 42 ...
##   ..- attr(*, "names")= chr [1:59] "-Andy Allanson" "-Billy Beane" "-Bruce Bochte" "-Bob Boone" ...
## NULL

建立完整的树:

tr1 <- tree(log(Salary) ~ Years + Hits, data=d)

剪枝为只有3个叶结点:

tr1b <- prune.tree(tr1, best=3)

显示树:

print(tr1b)
## node), split, n, deviance, yval
##       * denotes terminal node
## 
## 1) root 263 207.20 5.927  
##   2) Years < 4.5 90  42.35 5.107 *
##   3) Years > 4.5 173  72.71 6.354  
##     6) Hits < 117.5 90  28.09 5.998 *
##     7) Hits > 117.5 83  20.88 6.740 *

显示概括:

print(summary(tr1b))
## 
## Regression tree:
## snip.tree(tree = tr1, nodes = c(6L, 2L))
## Number of terminal nodes:  3 
## Residual mean deviance:  0.3513 = 91.33 / 260 
## Distribution of residuals:
##     Min.  1st Qu.   Median     Mean  3rd Qu.     Max. 
## -2.24000 -0.39580 -0.03162  0.00000  0.33380  2.55600

做树图:

plot(tr1b); text(tr1b, pretty=0)

38.2.5 树回归

把数据随机地分成一半训练集,一半测试集:

d <- na.omit(Hitters)
set.seed(1)
train <- sample(nrow(d), size=round(nrow(d)/2))
test <- (-train)

对训练集,建立未剪枝的树:

tr1 <- tree(log(Salary) ~ ., data=d, subset=train)
plot(tr1); text(tr1, pretty=0)
Hitters数据训练集未剪枝树

图38.6: Hitters数据训练集未剪枝树

对训练集上的未剪枝树用交叉验证方法寻找最优大小:

cv1 <- cv.tree(tr1)
print(cv1)
## $size
## [1] 8 7 6 5 4 3 2 1
## 
## $dev
## [1]  44.55223  44.45312  44.57906  44.53469  46.93001  54.03823  57.53660 105.17743
## 
## $k
## [1]      -Inf  1.679266  1.750440  1.836204  3.300858  6.230249  7.420672 56.727362
## 
## $method
## [1] "deviance"
## 
## attr(,"class")
## [1] "prune"         "tree.sequence"
plot(cv1$size, cv1$dev, type='b')
best.size <- cv1$size[which.min(cv1$dev)[1]]
abline(v=best.size, col='gray')

最优大小为7。 获得训练集上构造的树剪枝后的结果:

best.size <- 4
tr1b <- prune.tree(tr1, best=best.size)

在测试集上计算预测均方误差:

pred.test <- predict(tr1b, newdata=d[test,])
test.mse <- mean( (d[test, 'Salary'] - exp(pred.test))^2 )
test.mse
## [1] 128224.1

如果用训练集的因变量平均值估计测试集的因变量值, 均方误差为:

worst.mse <- mean( (d[test, 'Salary'] - mean(d[train, 'Salary']))^2 )
worst.mse
## [1] 224692.1

用所有数据来构造未剪枝树:

tr2 <- tree(log(Salary) ~ ., data=d)

用训练集上得到的子树大小剪枝:

tr2b <- prune.tree(tr2, best=best.size)
plot(tr2b); text(tr2b, pretty=0)

38.2.6 装袋法

判别树在不同的训练集、测试集划分上可以产生很大变化, 说明其预测值方差较大。 利用bootstrap的思想, 可以随机选取许多个训练集, 把许多个训练集的模型结果平均, 就可以降低预测值的方差。

办法是从一个训练集中用有放回抽样的方法抽取\(B\)个训练集, 设第\(b\)个抽取的训练集得到的回归函数为\(\hat f^{*b}(\cdot)\), 则最后的回归函数是这些回归函数的平均值: \[\begin{aligned} \hat f_{\text{bagging}}(x) = \frac{1}{B} \sum_{b=1}^b \hat f^{*b}(x) \end{aligned}\] 这称为装袋法(bagging)。 装袋法对改善判别与回归树的精度十分有效。

装袋法的步骤如下:

  • 从训练集中取\(B\)个有放回随机抽样的bootstrap训练集,\(B\)取为几百到几千之间。
  • 对每个bootstrap训练集,估计未剪枝的树。
  • 如果因变量是连续变量,对测试样品,用所有的树的预测值的平均值作预测。
  • 如果因变量是分类变量,对测试样品,可以用所有树预测类的多数投票决定预测值。

装袋法也可以用来改进其他的回归和判别方法。

装袋后不能再用图形表示,模型可解释性较差。 但是,可以度量自变量在预测中的重要程度。 在回归问题中, 可以计算每个自变量在所有\(B\)个树种平均减少的残差平方和的量, 以此度量其重要度。 在判别问题中, 可以计算每个自变量在所有\(B\)个树种平均减少的基尼系数的量, 以此度量其重要度。

除了可以用测试集、交叉验证方法以外, 还可以使用袋外观测预测误差。 用bootstrap再抽样获得多个训练集时每个bootstrap训练集总会遗漏一些观测, 平均每个bootstrap训练集会遗漏三分之一的观测。 对每个观测,大约有\(B/3\)棵树没有用到此观测, 可以用这些树的预测值平均来预测此观测,得到一个误差估计, 这样得到的均方误差估计或错判率称为袋外观测估计(OOB估计)。 好处是不用很多额外的工作。

对训练集用装袋法:

bag1 <- randomForest(log(Salary) ~ ., data=d, subset=train, mtry=ncol(d)-1, importance=TRUE)
bag1
## 
## Call:
##  randomForest(formula = log(Salary) ~ ., data = d, mtry = ncol(d) -      1, importance = TRUE, subset = train) 
##                Type of random forest: regression
##                      Number of trees: 500
## No. of variables tried at each split: 19
## 
##           Mean of squared residuals: 0.2549051
##                     % Var explained: 67.32

注意randomForest()函数实际是随机森林法, 但是当mtry的值取为所有自变量个数时就是装袋法。

对测试集进行预报:

pred2 <- predict(bag1, newdata=d[test,])
test.mse2 <- mean( (d[test, 'Salary'] - exp(pred2))^2 )
test.mse2
## [1] 89851.48

结果与剪枝过的单课树相近。

在全集上使用装袋法:

bag2 <- randomForest(log(Salary) ~ ., data=d, mtry=ncol(d)-1, importance=TRUE)
bag2
## 
## Call:
##  randomForest(formula = log(Salary) ~ ., data = d, mtry = ncol(d) -      1, importance = TRUE) 
##                Type of random forest: regression
##                      Number of trees: 500
## No. of variables tried at each split: 19
## 
##           Mean of squared residuals: 0.1894008
##                     % Var explained: 75.95

变量的重要度数值和图形: 各变量的重要度数值及其图形:

importance(bag2)
##              %IncMSE IncNodePurity
## AtBat     10.4186792     8.9315248
## Hits       8.0033436     7.5938472
## HmRun      3.6180992     1.9689157
## Runs       7.2586283     3.9341954
## RBI        5.9223739     5.9201328
## Walks      7.6449979     6.6988173
## Years      9.4732817     2.2977104
## CAtBat    28.2381456    84.4338845
## CHits     13.8405414    26.2455674
## CHmRun     6.7109109     3.8246805
## CRuns     13.6067783    29.1340423
## CRBI      14.1694017    10.9852537
## CWalks     7.7656943     4.1725799
## League    -1.0964577     0.2146176
## Division   0.5534057     0.2307037
## PutOuts    0.3157195     4.1169655
## Assists   -1.7730978     1.6599765
## Errors     2.3420783     1.6852796
## NewLeague -0.3091532     0.3747445
varImpPlot(bag2)
Hitters数据装袋法的变量重要性结果

图38.7: Hitters数据装袋法的变量重要性结果

最重要的自变量是CAtBats, 其次有CRuns, CHits等。

38.2.7 随机森林

随机森林的思想与装袋法类似, 但是试图使得参加平均的各个树之间变得比较独立。 仍采用有放回抽样得到的多个bootstrap训练集, 但是对每个bootstrap训练集构造判别树时, 每次分叉时不考虑所有自变量, 而是仅考虑随机选取的一个自变量子集。

对判别树,每次分叉时选取的自变量个数通常取\(m \approx \sqrt{p}\)个。 比如,对Heart数据的13个自变量, 每次分叉时仅随机选取4个纳入考察范围。

随机森林的想法是基于正相关的样本在平均时并不能很好地降低方差, 独立样本能比较好地降低方差。 如果存在一个最重要的变量, 如果不加限制这个最重要的变量总会是第一个分叉, 使得\(B\)棵树相似程度很高。 随机森林解决这个问题的办法是限制分叉时可选的变量子集。

随机森林也可以用来改进其他的回归和判别方法。

装袋法和随机森林都可以用R扩展包randomForest的 randomForest()函数实现。 当此函数的mtry参数取为自变量个数时,执行的就是装袋法; mtry取缺省值时,执行随机森林算法。 执行随机森林算法时, randomForest()函数在回归问题时分叉时考虑的自变量个数取\(m \approx p/3\), 在判别问题时取\(m \approx \sqrt{p}\)

对训练集用随机森林法:

rf1 <- randomForest(log(Salary) ~ ., data=d, subset=train, importance=TRUE)
rf1
## 
## Call:
##  randomForest(formula = log(Salary) ~ ., data = d, importance = TRUE,      subset = train) 
##                Type of random forest: regression
##                      Number of trees: 500
## No. of variables tried at each split: 6
## 
##           Mean of squared residuals: 0.2422914
##                     % Var explained: 68.94

mtry的值取为缺省值时执行随机森林算法。

对测试集进行预报:

pred3 <- predict(rf1, newdata=d[test,])
test.mse3 <- mean( (d[test, 'Salary'] - exp(pred3))^2 )
test.mse3
## [1] 95455.53

结果与剪枝过的单课树、装袋法相近。

在全集上使用随机森林:

rf2 <- randomForest(log(Salary) ~ ., data=d, importance=TRUE)
rf2
## 
## Call:
##  randomForest(formula = log(Salary) ~ ., data = d, importance = TRUE) 
##                Type of random forest: regression
##                      Number of trees: 500
## No. of variables tried at each split: 6
## 
##           Mean of squared residuals: 0.1789257
##                     % Var explained: 77.28

各变量的重要度数值及其图形:

importance(rf2)
##              %IncMSE IncNodePurity
## AtBat     10.1010786     7.6235571
## Hits       8.3614365     7.9201425
## HmRun      3.7354302     2.4553471
## Runs       7.9446786     4.6566217
## RBI        6.6246470     6.0251832
## Walks      8.8848565     5.9901840
## Years     11.3153391     8.4097999
## CAtBat    18.0377154    41.0851904
## CHits     17.5110322    37.8686798
## CHmRun     8.4476983     6.9078686
## CRuns     14.8793512    30.8423409
## CRBI      14.8308800    20.5339522
## CWalks     9.7578555    14.9467745
## League    -0.6619015     0.3041459
## Division  -0.3583808     0.2954341
## PutOuts    2.4366422     3.4901980
## Assists   -0.0965240     1.8105096
## Errors     1.5007791     1.7068960
## NewLeague  1.0593259     0.3453514
varImpPlot(rf2)
Hitters数据随机森林法的变量重要度结果

图38.8: Hitters数据随机森林法的变量重要度结果

最重要的自变量是CAtBats, CRuns, CHits, CWalks, CRBI等。

38.3 Heart数据分析

Heart数据是心脏病诊断的数据, 因变量AHD为是否有心脏病, 试图用各个自变量预测(判别)。

读入Heart数据集,并去掉有缺失值的观测:

Heart <- read.csv(
  "data/Heart.csv", header=TRUE, row.names=1,
  stringsAsFactors=TRUE)
Heart <- na.omit(Heart)
str(Heart)
## 'data.frame':    297 obs. of  14 variables:
##  $ Age      : int  63 67 67 37 41 56 62 57 63 53 ...
##  $ Sex      : int  1 1 1 1 0 1 0 0 1 1 ...
##  $ ChestPain: Factor w/ 4 levels "asymptomatic",..: 4 1 1 2 3 3 1 1 1 1 ...
##  $ RestBP   : int  145 160 120 130 130 120 140 120 130 140 ...
##  $ Chol     : int  233 286 229 250 204 236 268 354 254 203 ...
##  $ Fbs      : int  1 0 0 0 0 0 0 0 0 1 ...
##  $ RestECG  : int  2 2 2 0 2 0 2 0 2 2 ...
##  $ MaxHR    : int  150 108 129 187 172 178 160 163 147 155 ...
##  $ ExAng    : int  0 1 1 0 0 0 0 1 0 1 ...
##  $ Oldpeak  : num  2.3 1.5 2.6 3.5 1.4 0.8 3.6 0.6 1.4 3.1 ...
##  $ Slope    : int  3 2 2 3 1 1 3 1 2 3 ...
##  $ Ca       : int  0 3 2 0 0 0 2 0 1 0 ...
##  $ Thal     : Factor w/ 3 levels "fixed","normal",..: 1 2 3 2 2 2 2 2 3 3 ...
##  $ AHD      : Factor w/ 2 levels "No","Yes": 1 2 2 1 1 1 2 1 2 2 ...
##  - attr(*, "na.action")= 'omit' Named int [1:6] 88 167 193 267 288 303
##   ..- attr(*, "names")= chr [1:6] "88" "167" "193" "267" ...
t(summary(Heart))
##                                                                                                                               
##      Age         Min.   :29.00      1st Qu.:48.00      Median :56.00      Mean   :54.54      3rd Qu.:61.00    Max.   :77.00   
##      Sex         Min.   :0.0000     1st Qu.:0.0000     Median :1.0000     Mean   :0.6768     3rd Qu.:1.0000   Max.   :1.0000  
##        ChestPain asymptomatic:142   nonanginal  : 83   nontypical  : 49   typical     : 23                                    
##     RestBP       Min.   : 94.0      1st Qu.:120.0      Median :130.0      Mean   :131.7      3rd Qu.:140.0    Max.   :200.0   
##      Chol        Min.   :126.0      1st Qu.:211.0      Median :243.0      Mean   :247.4      3rd Qu.:276.0    Max.   :564.0   
##      Fbs         Min.   :0.0000     1st Qu.:0.0000     Median :0.0000     Mean   :0.1448     3rd Qu.:0.0000   Max.   :1.0000  
##    RestECG       Min.   :0.0000     1st Qu.:0.0000     Median :1.0000     Mean   :0.9966     3rd Qu.:2.0000   Max.   :2.0000  
##     MaxHR        Min.   : 71.0      1st Qu.:133.0      Median :153.0      Mean   :149.6      3rd Qu.:166.0    Max.   :202.0   
##     ExAng        Min.   :0.0000     1st Qu.:0.0000     Median :0.0000     Mean   :0.3266     3rd Qu.:1.0000   Max.   :1.0000  
##    Oldpeak       Min.   :0.000      1st Qu.:0.000      Median :0.800      Mean   :1.056      3rd Qu.:1.600    Max.   :6.200   
##     Slope        Min.   :1.000      1st Qu.:1.000      Median :2.000      Mean   :1.603      3rd Qu.:2.000    Max.   :3.000   
##       Ca         Min.   :0.0000     1st Qu.:0.0000     Median :0.0000     Mean   :0.6768     3rd Qu.:1.0000   Max.   :3.0000  
##         Thal     fixed     : 18     normal    :164     reversable:115                                                         
##  AHD             No :160            Yes:137

数据下载:Heart.csv

38.3.1 树回归

38.3.1.1 划分训练集与测试集

简单地把观测分为一半训练集、一半测试集:

set.seed(1)
train <- sample(nrow(Heart), size=round(nrow(Heart)/2))
test <- (-train)
test.y <- Heart[test, 'AHD']

在训练集上建立未剪枝的判别树:

tr1 <- tree(AHD ~ ., data=Heart[train,])
plot(tr1); text(tr1, pretty=0)

38.3.1.2 适当剪枝

用交叉验证方法确定剪枝保留的叶子个数,剪枝时按照错判率执行:

cv1 <- cv.tree(tr1, FUN=prune.misclass)
cv1
## $size
## [1] 12  9  6  4  2  1
## 
## $dev
## [1] 42 44 47 44 57 69
## 
## $k
## [1]      -Inf  0.000000  1.666667  3.000000  7.000000 26.000000
## 
## $method
## [1] "misclass"
## 
## attr(,"class")
## [1] "prune"         "tree.sequence"
plot(cv1$size, cv1$dev, type='b', xlab='size', ylab='dev')

best.size <- cv1$size[which.min(cv1$dev)]

最优的大小是12。但是从图上看,4个叶结点已经足够好,所以取为4。

对训练集生成剪枝结果:

best.size <- 4
tr1b <- prune.misclass(tr1, best=best.size)
plot(tr1b); text(tr1b, pretty=0)
Heart数据回归树

图38.9: Heart数据回归树

注意剪枝后树的显示中, 内部节点的自变量存在分类变量, 这时按照这个自变量分叉时, 取指定的某几个分类值时对应分支Yes, 取其它的分类值时对应分支No。

38.3.1.3 对测试集计算误判率

pred1 <- predict(tr1b, Heart[test,], type='class')
tab1 <- table(pred1, test.y); tab1
##      test.y
## pred1 No Yes
##   No  56  17
##   Yes 21  55
test.err <- (tab1[1,2]+tab1[2,1])/sum(tab1[]); test.err
## [1] 0.2550336

对测试集的错判率约26%。

利用未剪枝的树对测试集进行预测, 一般比剪枝后的结果差:

pred1a <- predict(tr1, Heart[test,], type='class')
tab1a <- table(pred1a, test.y); tab1a
##       test.y
## pred1a No Yes
##    No  58  21
##    Yes 19  51
test.err1a <- (tab1a[1,2]+tab1a[2,1])/sum(tab1a[]); test.err1a
## [1] 0.2684564

38.3.1.4 利用全集数据建立剪枝判别树

tr2 <- tree(AHD ~ ., data=Heart)
tr2b <- prune.misclass(tr2, best=best.size)
plot(tr2b); text(tr2b, pretty=0)

38.3.2 用装袋法

对训练集用装袋法:

bag1 <- randomForest(AHD ~ ., data=Heart, subset=train, mtry=13, importance=TRUE)
bag1
## 
## Call:
##  randomForest(formula = AHD ~ ., data = Heart, mtry = 13, importance = TRUE,      subset = train) 
##                Type of random forest: classification
##                      Number of trees: 500
## No. of variables tried at each split: 13
## 
##         OOB estimate of  error rate: 22.3%
## Confusion matrix:
##     No Yes class.error
## No  71  12   0.1445783
## Yes 21  44   0.3230769

注意randomForest()函数实际是随机森林法, 但是当mtry的值取为所有自变量个数时就是装袋法。 袋外观测得到的错判率比较差。

对测试集进行预报:

pred2 <- predict(bag1, newdata=Heart[test,])
tab2 <- table(pred2, test.y); tab2
##      test.y
## pred2 No Yes
##   No  66  17
##   Yes 11  55
test.err2 <- (tab2[1,2]+tab2[2,1])/sum(tab2[]); test.err2
## [1] 0.1879195

测试集的错判率约为19%。

对全集用装袋法:

bag1b <- randomForest(AHD ~ ., data=Heart, mtry=13, importance=TRUE)
bag1b
## 
## Call:
##  randomForest(formula = AHD ~ ., data = Heart, mtry = 13, importance = TRUE) 
##                Type of random forest: classification
##                      Number of trees: 500
## No. of variables tried at each split: 13
## 
##         OOB estimate of  error rate: 20.88%
## Confusion matrix:
##      No Yes class.error
## No  131  29   0.1812500
## Yes  33 104   0.2408759

各变量的重要度数值及其图形:

importance(bag1b)
##                   No         Yes MeanDecreaseAccuracy MeanDecreaseGini
## Age        6.5766876  5.12005531            8.7542379       12.2956568
## Sex       11.2077275  4.48390165           11.2853739        3.7278623
## ChestPain 13.0268932 17.89348038           20.4292863       23.3424850
## RestBP     2.6203153  0.05626759            2.0521195        9.7650173
## Chol      -0.8712348 -4.23294461           -3.0733270       11.5911988
## Fbs       -0.6941335 -1.16860850           -1.2288380        0.6775051
## RestECG   -1.4881617  0.23292163           -0.8772267        1.8426038
## MaxHR      7.7625054  2.34660468            7.5122314       13.2101707
## ExAng      2.7926364  5.45108497            5.7525854        3.5491718
## Oldpeak   14.8193517 14.67748373           20.2425364       14.5480191
## Slope      2.5189935  5.73789018            5.9744484        4.2777028
## Ca        23.0513399 18.01671793           27.4320740       20.0564750
## Thal      20.1968435 18.74418431           25.0618361       28.2479833
varImpPlot(bag1b)

最重要的变量是Thal, ChestPain, Ca。

38.3.3 用随机森林

对训练集用随机森林法:

rf1 <- randomForest(AHD ~ ., data=Heart, subset=train, importance=TRUE)
rf1
## 
## Call:
##  randomForest(formula = AHD ~ ., data = Heart, importance = TRUE,      subset = train) 
##                Type of random forest: classification
##                      Number of trees: 500
## No. of variables tried at each split: 3
## 
##         OOB estimate of  error rate: 21.62%
## Confusion matrix:
##     No Yes class.error
## No  71  12   0.1445783
## Yes 20  45   0.3076923

这里mtry取缺省值,对应于随机森林法。

对测试集进行预报:

pred3 <- predict(rf1, newdata=Heart[test,])
tab3 <- table(pred3, test.y); tab3
##      test.y
## pred3 No Yes
##   No  70  16
##   Yes  7  56
test.err3 <- (tab3[1,2]+tab3[2,1])/sum(tab3[]); test.err3
## [1] 0.1543624

测试集的错判率约为15%。

对全集用随机森林:

rf1b <- randomForest(AHD ~ ., data=Heart,  importance=TRUE)
rf1b
## 
## Call:
##  randomForest(formula = AHD ~ ., data = Heart, importance = TRUE) 
##                Type of random forest: classification
##                      Number of trees: 500
## No. of variables tried at each split: 3
## 
##         OOB estimate of  error rate: 16.5%
## Confusion matrix:
##      No Yes class.error
## No  140  20   0.1250000
## Yes  29 108   0.2116788

各变量的重要度数值及其图形:

importance(rf1b)
##                   No        Yes MeanDecreaseAccuracy MeanDecreaseGini
## Age        7.2380857  5.4451404            9.1859647        12.908917
## Sex       10.1973138  8.0790483           12.6929315         4.938266
## ChestPain 10.4623927 16.7054395           18.8771946        18.218363
## RestBP     1.2157266  1.8875229            2.1025511        10.624864
## Chol      -1.2630538 -0.4285615           -1.3028275        11.470420
## Fbs        0.4417651 -2.6574949           -1.4327524         1.418137
## RestECG   -1.1149040  1.4649476            0.2220661         2.840670
## MaxHR      9.3788412  6.0542618           10.7139921        17.383623
## ExAng      3.3923281  9.8037831            9.3523828         6.715947
## Oldpeak   10.1617047 14.3372404           17.5616061        15.403935
## Slope      2.6703016  9.3147738            8.5774100         6.752552
## Ca        21.1750038 20.2285033           26.7114362        18.388524
## Thal      18.0250446 16.6731737           22.5365589        18.351175
varImpPlot(rf1b)
Heart数据随机森林方法得到的变量重要度

图38.10: Heart数据随机森林方法得到的变量重要度

最重要的变量是ChestPain, Thal, Ca。

38.4 汽车销量数据分析

Carseats是ISLR包的一个数据集,基本情况如下: {rstatl-car-summ01, cache=TRUE} str(Carseats) summary(Carseats)

把Salses变量按照大于8与否分成两组, 结果存入变量High,以High为因变量作判别分析。

d <- na.omit(Carseats)
d$High <- factor(ifelse(d$Sales > 8, 'Yes', 'No'))
dim(d)
## [1] 400  12

38.4.1 判别树

38.4.1.1 全体数据的判别树

对全体数据建立未剪枝的判别树:

tr1 <- tree(High ~ . - Sales, data=d)
summary(tr1)
## 
## Classification tree:
## tree(formula = High ~ . - Sales, data = d)
## Variables actually used in tree construction:
## [1] "ShelveLoc"   "Price"       "Income"      "CompPrice"   "Population"  "Advertising" "Age"         "US"         
## Number of terminal nodes:  27 
## Residual mean deviance:  0.4575 = 170.7 / 373 
## Misclassification error rate: 0.09 = 36 / 400
plot(tr1)
text(tr1, pretty=0)

38.4.1.2 划分训练集和测试集

把输入数据集随机地分一半当作训练集,另一半当作测试集:

set.seed(2)
train <- sample(nrow(d), size=round(nrow(d)/2))
test <- (-train)
test.high <- d[test, 'High']

用训练数据建立未剪枝的判别树:

tr2 <- tree(High ~ . - Sales, data=d, subset=train)
summary(tr2)
## 
## Classification tree:
## tree(formula = High ~ . - Sales, data = d, subset = train)
## Variables actually used in tree construction:
## [1] "Price"       "Population"  "ShelveLoc"   "Age"         "Education"   "CompPrice"   "Advertising" "Income"      "US"         
## Number of terminal nodes:  21 
## Residual mean deviance:  0.5543 = 99.22 / 179 
## Misclassification error rate: 0.115 = 23 / 200
plot(tr2)
text(tr2, pretty=0)

用未剪枝的树对测试集进行预测,并计算误判率:

pred2 <- predict(tr2, d[test,], type='class')
tab <- table(pred2, test.high); tab
##      test.high
## pred2  No Yes
##   No  104  33
##   Yes  13  50
test.err2 <- (tab[1,2] + tab[2,1]) / sum(tab[]); test.err2
## [1] 0.23

38.4.1.3 用交叉验证确定训练集的剪枝

set.seed(3)
cv1 <- cv.tree(tr2, FUN=prune.misclass)
cv1
## $size
## [1] 21 19 14  9  8  5  3  2  1
## 
## $dev
## [1] 74 76 81 81 75 77 78 85 81
## 
## $k
## [1] -Inf  0.0  1.0  1.4  2.0  3.0  4.0  9.0 18.0
## 
## $method
## [1] "misclass"
## 
## attr(,"class")
## [1] "prune"         "tree.sequence"
plot(cv1$size, cv1$dev, type='b')

best.size <- cv1$size[which.min(cv1$dev)]

用交叉验证方法自动选择的最佳树大小为21。

剪枝:

tr3 <- prune.misclass(tr2, best=best.size)
summary(tr3)
## 
## Classification tree:
## tree(formula = High ~ . - Sales, data = d, subset = train)
## Variables actually used in tree construction:
## [1] "Price"       "Population"  "ShelveLoc"   "Age"         "Education"   "CompPrice"   "Advertising" "Income"      "US"         
## Number of terminal nodes:  21 
## Residual mean deviance:  0.5543 = 99.22 / 179 
## Misclassification error rate: 0.115 = 23 / 200
plot(tr3)
text(tr3, pretty=0)

用剪枝后的树对测试集进行预测,计算误判率:

pred3 <- predict(tr3, d[test,], type='class')
tab <- table(pred3, test.high); tab
##      test.high
## pred3  No Yes
##   No  104  32
##   Yes  13  51
test.err3 <- (tab[1,2] + tab[2,1]) / sum(tab[]); test.err3
## [1] 0.225

38.4.2 随机森林

对训练集用随机森林法:

rf4 <- randomForest(High ~ . - Sales, data=d, subset=train, importance=TRUE)
rf4
## 
## Call:
##  randomForest(formula = High ~ . - Sales, data = d, importance = TRUE,      subset = train) 
##                Type of random forest: classification
##                      Number of trees: 500
## No. of variables tried at each split: 3
## 
##         OOB estimate of  error rate: 25.5%
## Confusion matrix:
##      No Yes class.error
## No  102  17   0.1428571
## Yes  34  47   0.4197531

这里mtry取缺省值,对应于随机森林法。

对测试集进行预报:

pred4 <- predict(rf4, newdata=d[test,])
tab <- table(pred4, test.high); tab
##      test.high
## pred4  No Yes
##   No  109  24
##   Yes   8  59
test.err4 <- (tab[1,2]+tab[2,1])/sum(tab[]); test.err4
## [1] 0.16

注意错判率结果依赖于训练集和测试集的划分, 另行选择训练集与测试集可能会得到很不一样的错判率结果。

对全集用随机森林:

rf5 <- randomForest(High ~ . - Sales, data=d,  importance=TRUE)
rf5
## 
## Call:
##  randomForest(formula = High ~ . - Sales, data = d, importance = TRUE) 
##                Type of random forest: classification
##                      Number of trees: 500
## No. of variables tried at each split: 3
## 
##         OOB estimate of  error rate: 18.25%
## Confusion matrix:
##      No Yes class.error
## No  213  23  0.09745763
## Yes  50 114  0.30487805

各变量的重要度数值及其图形:

importance(rf5)
##                     No        Yes MeanDecreaseAccuracy MeanDecreaseGini
## CompPrice   11.0998129  5.4168875          11.65469579        21.820876
## Income       3.2897388  4.7177705           5.75865440        20.384692
## Advertising 10.8093624 16.0263308          18.31175988        23.350563
## Population  -3.1872660 -1.7367798          -3.63402082        15.670307
## Price       30.0864270 28.7929995          37.44125656        43.492787
## ShelveLoc   30.2789749 33.8109594          39.67983055        30.053785
## Age          9.7116826  9.0261373          12.78808426        22.578000
## Education    0.2214031 -0.3203644           0.06365633         9.899447
## Urban        1.3826674  1.4199879           1.98859615         2.128048
## US           3.7289827  5.1909662           6.83788775         3.405420
varImpPlot(rf5)
Carseats数据随机森林法得到的变量重要度

图38.11: Carseats数据随机森林法得到的变量重要度

重要的自变量为Price, ShelfLoc, 其次有Age, Advertising, CompPrice, Income等。

38.5 波士顿郊区房价数据

MASS包的Boston数据包含了波士顿地区郊区房价的若干数据。 以中位房价medv为因变量建立回归模型。 首先把缺失值去掉后存入数据集d:

d <- na.omit(Boston)

数据集概况:

str(d)
## 'data.frame':    506 obs. of  14 variables:
##  $ crim   : num  0.00632 0.02731 0.02729 0.03237 0.06905 ...
##  $ zn     : num  18 0 0 0 0 0 12.5 12.5 12.5 12.5 ...
##  $ indus  : num  2.31 7.07 7.07 2.18 2.18 2.18 7.87 7.87 7.87 7.87 ...
##  $ chas   : int  0 0 0 0 0 0 0 0 0 0 ...
##  $ nox    : num  0.538 0.469 0.469 0.458 0.458 0.458 0.524 0.524 0.524 0.524 ...
##  $ rm     : num  6.58 6.42 7.18 7 7.15 ...
##  $ age    : num  65.2 78.9 61.1 45.8 54.2 58.7 66.6 96.1 100 85.9 ...
##  $ dis    : num  4.09 4.97 4.97 6.06 6.06 ...
##  $ rad    : int  1 2 2 3 3 3 5 5 5 5 ...
##  $ tax    : num  296 242 242 222 222 222 311 311 311 311 ...
##  $ ptratio: num  15.3 17.8 17.8 18.7 18.7 18.7 15.2 15.2 15.2 15.2 ...
##  $ black  : num  397 397 393 395 397 ...
##  $ lstat  : num  4.98 9.14 4.03 2.94 5.33 ...
##  $ medv   : num  24 21.6 34.7 33.4 36.2 28.7 22.9 27.1 16.5 18.9 ...
summary(d)
##       crim                zn             indus            chas              nox               rm             age              dis              rad              tax           ptratio          black            lstat            medv      
##  Min.   : 0.00632   Min.   :  0.00   Min.   : 0.46   Min.   :0.00000   Min.   :0.3850   Min.   :3.561   Min.   :  2.90   Min.   : 1.130   Min.   : 1.000   Min.   :187.0   Min.   :12.60   Min.   :  0.32   Min.   : 1.73   Min.   : 5.00  
##  1st Qu.: 0.08205   1st Qu.:  0.00   1st Qu.: 5.19   1st Qu.:0.00000   1st Qu.:0.4490   1st Qu.:5.886   1st Qu.: 45.02   1st Qu.: 2.100   1st Qu.: 4.000   1st Qu.:279.0   1st Qu.:17.40   1st Qu.:375.38   1st Qu.: 6.95   1st Qu.:17.02  
##  Median : 0.25651   Median :  0.00   Median : 9.69   Median :0.00000   Median :0.5380   Median :6.208   Median : 77.50   Median : 3.207   Median : 5.000   Median :330.0   Median :19.05   Median :391.44   Median :11.36   Median :21.20  
##  Mean   : 3.61352   Mean   : 11.36   Mean   :11.14   Mean   :0.06917   Mean   :0.5547   Mean   :6.285   Mean   : 68.57   Mean   : 3.795   Mean   : 9.549   Mean   :408.2   Mean   :18.46   Mean   :356.67   Mean   :12.65   Mean   :22.53  
##  3rd Qu.: 3.67708   3rd Qu.: 12.50   3rd Qu.:18.10   3rd Qu.:0.00000   3rd Qu.:0.6240   3rd Qu.:6.623   3rd Qu.: 94.08   3rd Qu.: 5.188   3rd Qu.:24.000   3rd Qu.:666.0   3rd Qu.:20.20   3rd Qu.:396.23   3rd Qu.:16.95   3rd Qu.:25.00  
##  Max.   :88.97620   Max.   :100.00   Max.   :27.74   Max.   :1.00000   Max.   :0.8710   Max.   :8.780   Max.   :100.00   Max.   :12.127   Max.   :24.000   Max.   :711.0   Max.   :22.00   Max.   :396.90   Max.   :37.97   Max.   :50.00

38.5.1 回归树

38.5.1.1 划分训练集和测试集

set.seed(1)
train <- sample(nrow(d), size=round(nrow(d)/2))
test <- (-train)

对训练集建立未剪枝的树:

tr1 <- tree(medv ~ ., d, subset=train)
summary(tr1)
## 
## Regression tree:
## tree(formula = medv ~ ., data = d, subset = train)
## Variables actually used in tree construction:
## [1] "rm"    "lstat" "crim"  "age"  
## Number of terminal nodes:  7 
## Residual mean deviance:  10.38 = 2555 / 246 
## Distribution of residuals:
##     Min.  1st Qu.   Median     Mean  3rd Qu.     Max. 
## -10.1800  -1.7770  -0.1775   0.0000   1.9230  16.5800
plot(tr1)
text(tr1, pretty=0)

用未剪枝的树对测试集进行预测,计算均方误差:

yhat <-predict(tr1, newdata=d[test,])
mse1 <- mean((yhat - d[test, 'medv'])^2)
mse1
## [1] 35.28688

38.5.1.2 用交叉验证方法确定剪枝复杂度

cv1 <- cv.tree(tr1)
plot(cv1$size, cv1$dev, type='b')

best.size <- cv1$size[which.min(cv1$dev)]; best.size
## [1] 7

剪枝并对测试集进行预测:

tr2 <- prune.tree(tr1, best=best.size)
plot(tr2)
text(tr2, pretty=0)

yhat <-predict(tr2, newdata=d[test,])
mse2 <- mean((yhat - d[test, 'medv'])^2)
mse2
## [1] 35.28688

剪枝后效果没有改善。

38.5.2 装袋法

用randomForest包计算。 当参数mtry取为自变量个数时按照装袋法计算。 对训练集计算。

set.seed(1)
bag1 <- randomForest(
  medv ~ ., data=d, subset=train, 
  mtry=ncol(d)-1, importance=TRUE)
bag1
## 
## Call:
##  randomForest(formula = medv ~ ., data = d, mtry = ncol(d) - 1,      importance = TRUE, subset = train) 
##                Type of random forest: regression
##                      Number of trees: 500
## No. of variables tried at each split: 13
## 
##           Mean of squared residuals: 11.39601
##                     % Var explained: 85.17

在测试集上计算装袋法的均方误差:

yhat <- predict(bag1, newdata=d[test,])
mean( (yhat - d[test, 'medv'])^2 )
## [1] 23.59273

比单棵树的结果有明显改善。

38.5.3 随机森林

用randomForest包计算。 当参数mtry取为缺省值时按照随机森林方法计算。 对训练集计算。

set.seed(1)
rf1 <- randomForest(
  medv ~ ., data=d, subset=train, 
  importance=TRUE)
rf1
## 
## Call:
##  randomForest(formula = medv ~ ., data = d, importance = TRUE,      subset = train) 
##                Type of random forest: regression
##                      Number of trees: 500
## No. of variables tried at each split: 4
## 
##           Mean of squared residuals: 10.23441
##                     % Var explained: 86.69

在测试集上计算随机森林法的均方误差:

yhat <- predict(rf1, newdata=d[test,])
mean( (yhat - d[test, 'medv'])^2 )
## [1] 18.11686

比单棵树的结果有明显改善, 比装袋法的结果也好一些。

各变量的重要度数值及其图形:

importance(rf1)
##           %IncMSE IncNodePurity
## crim    15.372334    1220.14856
## zn       3.335435     194.85945
## indus    6.964559    1021.94751
## chas     2.059298      69.68099
## nox     14.009761    1005.14707
## rm      28.693900    6162.30720
## age     13.832143     708.55138
## dis     10.317731     852.33701
## rad      4.390624     162.22597
## tax      7.536563     564.60422
## ptratio  9.333716    1163.39624
## black    8.341316     355.62445
## lstat   27.132450    5549.25088
varImpPlot(rf1)
Boston数据用随机森林法得到的变量重要度

图38.12: Boston数据用随机森林法得到的变量重要度

38.5.4 提升法

提升法(Boosting)也是可以用在多种回归和判别问题中的方法。 提升法的想法是,用比较简单的模型拟合因变量, 计算残差, 然后以残差为新的因变量建模, 仍使用简单的模型, 把两次的回归函数作加权和, 得到新的残差后,再以新残差作为因变量建模, 如此重复地更新回归函数, 得到由多个回归函数加权和组成的最终的回归函数。

加权一般取为比较小的值, 其目的是降低逼近速度。 统计学习问题中降低逼近速度一般结果更好。

提升法算法:

  • [(1)] 对训练集,设置\(r_i = y_i\),并令初始回归函数为\(\hat f(\cdot)=0\)

  • [(2)] 对\(b=1,2,\dots,B\)重复执行:

    • [(a)] 以训练集的自变量为自变量,以\(r\)为因变量,拟合一个仅有\(d\)个分叉的简单树回归函数, 设为\(\hat f_b\)
    • [(b)] 更新回归函数,添加一个压缩过的树回归函数: \[\begin{aligned} \hat f(x) \leftarrow \hat f(x) + \lambda \hat f_b(x); \end{aligned}\]
    • [(c)] 更新残差: \[\begin{aligned} r_i \leftarrow r_i - \lambda \hat f_b(x_i). \end{aligned}\]
  • [(3)] 提升法的回归函数为 \[\begin{aligned} \hat f(x) = \sum_{b=1}^B \lambda \hat f_b(x) . \end{aligned}\]

用多少个回归函数做加权和,即\(B\)的选取问题。 取得\(B\)太大也会有过度拟合, 但是只要\(B\)不太大这个问题不严重。 可以用交叉验证选择\(B\)的值。

收缩系数\(\lambda\)。 是一个小的正数, 控制学习速度, 经常用0.01, 0.001这样的值, 与要解决的问题有关。 取\(\lambda\)很小,就需要取\(B\)很大。

用来控制每个回归函数复杂度的参数, 对树回归而言就是树的大小。 一个分叉的树往往就很好。 取单个分叉时结果模型是可加模型, 没有交互项, 这是因为每个加权相加得回归函数都只依赖于单一自变量。 \(d>1\)时就加入了交互项。

使用gbm包。 在训练集上拟合:

set.seed(1)
bst1 <- gbm(medv ~ ., data=d[train,],  distribution='gaussian',  n.trees=5000,  interaction.depth=4)
summary(bst1)

##             var    rel.inf
## rm           rm 43.9919329
## lstat     lstat 33.1216941
## crim       crim  4.2604167
## dis         dis  4.0111090
## nox         nox  3.4353017
## black     black  2.8267554
## age         age  2.6113938
## ptratio ptratio  2.5403035
## tax         tax  1.4565654
## indus     indus  0.8008740
## rad         rad  0.6546400
## zn           zn  0.1446149
## chas       chas  0.1443986

lstat和rm是最重要的变量。

在测试集上预报,并计算均方误差:

yhat <- predict(bst1, newdata=d[test,], n.trees=5000)
mean( (yhat - d[test, 'medv'])^2 )
## [1] 18.84709

与随机森林方法结果相近。

如果提高学习速度:

bst2 <- gbm(medv ~ ., data=d[train,],  distribution='gaussian',  n.trees=5000,  interaction.depth=4, shrinkage=0.2)
yhat <- predict(bst2, newdata=d[test,], n.trees=5000)
mean( (yhat - d[test, 'medv'])^2 )
## [1] 18.33455

均方误差有改善。

38.6 支持向量机方法

支持向量机是1990年代有计算机科学家发明的一种有监督学习方法, 使用范围较广,预测精度较高。

支持向量机利用了Hilbert空间的方法将线性问题扩展为非线性问题。 线性的支持向量判别法, 可以通过\(\mathbb R^p\)的内积将线性的判别函数转化为如下的表示:

\[\begin{aligned} f(\boldsymbol x) = \beta_0 + \sum_{i=1}^n \alpha_i \langle \boldsymbol x, \boldsymbol x_i \rangle \end{aligned}\] 其中\(\beta_0, \alpha_1, \dots, \alpha_n\)是待定参数。 为了估计参数, 不需要用到各\(\boldsymbol x_i\)的具体值, 而只需要其两两的内积值, 而且在判别函数中只有支持向量对应的\(\alpha_i\)才非零, 记\(\mathcal S\)为支持向量点集, 则线性判别函数为 \[\begin{aligned} f(\boldsymbol x) = \beta_0 + \sum_{i \in \mathcal S} \alpha_i \langle \boldsymbol x, \boldsymbol x_i \rangle \end{aligned}\]

支持向量机方法将\(\mathbb R^p\)中的内积推广为如下的核函数值: \[\begin{aligned} K(\boldsymbol x, \boldsymbol x') \end{aligned}\] 核函数\(K(\boldsymbol x, \boldsymbol x')\), \(\boldsymbol x, \boldsymbol x' \in \mathbb R^p\) 是度量两个观测点\(\boldsymbol x, \boldsymbol x'\)的相似程度的函数。 比如, 取 \[\begin{aligned} K(\boldsymbol x, \boldsymbol x') = \sum_{j=1}^p x_j x_j' \end{aligned}\] 就又回到了线性的支持向量判别法。

核有多种取法。 例如, 取 \[\begin{aligned} K(\boldsymbol x, \boldsymbol x') = \left\{ 1 + \sum_{j=1}^p x_j x_j' \right\}^d \end{aligned}\] 其中\(d>1\)为正整数, 称为多项式核, 则结果是多项式边界的判别法, 本质上是对线性的支持向量方法添加了高次项和交叉项。

利用核代替内积后, 判别法的判别函数变成 \[\begin{aligned} f(\boldsymbol x) = \beta_0 + \sum_{i \in \mathcal S} K(\boldsymbol x, \boldsymbol x_i) \end{aligned}\]

另一种常用的核是径向核(radial kernel), 定义为 \[\begin{aligned} K(\boldsymbol x, \boldsymbol x') = \exp\left\{ - \gamma \sum_{j=1}^p (x_j - x_j')^2 \right\} \end{aligned}\] \(\gamma\)为正常数。 当\(\boldsymbol x\)\(\boldsymbol x'\)分别落在以原点为中心的两个超球面上时, 其核函数值不变。

使用径向核时, 判别函数为 \[\begin{aligned} f(\boldsymbol x) = \beta_0 + \sum_{i \in \mathcal S} \exp\left\{ - \gamma \sum_{j=1}^p (x_{j} - x_{ij})^2 \right\} \end{aligned}\] 对一个待判别的观测\(\boldsymbol x^*\), 如果\(\boldsymbol x^*\)距离训练观测点\(\boldsymbol x_i\)较远, 则\(K(\boldsymbol x^*, \boldsymbol x_i)\)的值很小, \(\boldsymbol x_i\)\(\boldsymbol x^*\)的判别基本不起作用。 这样的性质使得径向核方法具有很强的局部性, 只有离\(\boldsymbol x^*\)很近的点才对其判别起作用。

为什么采用核函数计算观测两两的\(\binom{n}{2}\)个核函数值, 而不是直接增加非线性项? 原因是计算这些核函数值计算量是确定的, 而增加许多非线性项, 则可能有很大的计算量, 而且某些核如径向核对应的自变量空间维数是无穷维的, 不能通过添加维度的办法解决。

支持向量机的理论基于再生核希尔伯特空间(RKHS), 可参见(Trevor Hastie 2009)节5.8和节12.3.3。

38.6.1 支持向量机用于Heart数据

考虑心脏病数据Heart的判别。 共297个观测, 随机选取其中207个作为训练集, 90个作为测试集。

set.seed(1)
Heart <- read.csv(
  "data/Heart.csv", header=TRUE, row.names=1,
  stringsAsFactors=TRUE)
d <- na.omit(Heart)
train <- sample(nrow(d), size=207)
test <- -train
d[["AHD"]] <- factor(d[["AHD"]], levels=c("No", "Yes"))

定义一个错判率函数:

classifier.error <- function(truth, pred){
  tab1 <- table(truth, pred)
  err <- 1 - sum(diag(tab1))/sum(c(tab1))
  err
}

38.6.1.1 线性的SVM

支持向量判别法就是SVM取多项式核, 阶数\(d=1\)的情形。 需要一个调节参数costcost越大, 分隔边界越窄, 过度拟合危险越大。

先随便取调节参数cost=1试验支持向量判别法:

res.svc <- svm(AHD ~ ., data=d[train,], kernel="linear", cost=1, scale=TRUE)
fit.svc <- predict(res.svc)
summary(res.svc)
## 
## Call:
## svm(formula = AHD ~ ., data = d[train, ], kernel = "linear", cost = 1, scale = TRUE)
## 
## 
## Parameters:
##    SVM-Type:  C-classification 
##  SVM-Kernel:  linear 
##        cost:  1 
## 
## Number of Support Vectors:  79
## 
##  ( 38 41 )
## 
## 
## Number of Classes:  2 
## 
## Levels: 
##  No Yes

计算拟合结果并计算错判率:

tab1 <- table(truth=d[train,"AHD"], fitted=fit.svc); tab1
##      fitted
## truth  No Yes
##   No  105   9
##   Yes  18  75
cat("SVC错判率:", round((tab1[1,2] + tab1[2,1])/ sum(c(tab1)), 2), "\n")
## SVC错判率: 0.13

e1071函数提供了tune()函数, 可以在训练集上用十折交叉验证选择较好的调节参数。

set.seed(101)
res.tune <- tune(svm, AHD ~ ., data=d[train,], kernel="linear", scale=TRUE,
                 ranges=list(cost=c(0.001, 0.01, 0.1, 1, 5, 10, 100, 1000)))
summary(res.tune)
## 
## Parameter tuning of 'svm':
## 
## - sampling method: 10-fold cross validation 
## 
## - best parameters:
##  cost
##   0.1
## 
## - best performance: 0.1542857 
## 
## - Detailed performance results:
##    cost     error dispersion
## 1 1e-03 0.4450000 0.08509809
## 2 1e-02 0.1695238 0.07062868
## 3 1e-01 0.1542857 0.07006458
## 4 1e+00 0.1590476 0.07793796
## 5 5e+00 0.1590476 0.08709789
## 6 1e+01 0.1590476 0.08709789
## 7 1e+02 0.1590476 0.08709789
## 8 1e+03 0.1590476 0.08709789

找到的最优调节参数为0.1, 可以用res.tune$best.model获得对应于最优调节参数的模型:

summary(res.tune$best.model)
## 
## Call:
## best.tune(method = svm, train.x = AHD ~ ., data = d[train, ], ranges = list(cost = c(0.001, 0.01, 0.1, 1, 5, 10, 100, 1000)), kernel = "linear", scale = TRUE)
## 
## 
## Parameters:
##    SVM-Type:  C-classification 
##  SVM-Kernel:  linear 
##        cost:  0.1 
## 
## Number of Support Vectors:  90
## 
##  ( 44 46 )
## 
## 
## Number of Classes:  2 
## 
## Levels: 
##  No Yes

在测试集上测试:

pred.svc <- predict(res.tune$best.model, newdata=d[test,])
tab1 <- table(truth=d[test,"AHD"], predict=pred.svc); tab1
##      predict
## truth No Yes
##   No  43   3
##   Yes 11  33
cat("SVC错判率:", round((tab1[1,2] + tab1[2,1])/ sum(c(tab1)), 2), "\n")
## SVC错判率: 0.16

38.6.1.2 多项式核SVM

res.svm1 <- svm(AHD ~ ., data=d[train,], kernel="polynomial", 
                order=2, cost=0.1, scale=TRUE)
fit.svm1 <- predict(res.svm1)
summary(res.svm1)
## 
## Call:
## svm(formula = AHD ~ ., data = d[train, ], kernel = "polynomial", order = 2, cost = 0.1, scale = TRUE)
## 
## 
## Parameters:
##    SVM-Type:  C-classification 
##  SVM-Kernel:  polynomial 
##        cost:  0.1 
##      degree:  3 
##      coef.0:  0 
## 
## Number of Support Vectors:  187
## 
##  ( 92 95 )
## 
## 
## Number of Classes:  2 
## 
## Levels: 
##  No Yes
tab1 <- table(truth=d[train,"AHD"], fitted=fit.svm1); tab1
##      fitted
## truth  No Yes
##   No  114   0
##   Yes  82  11
cat("2阶多项式核SVM错判率:", round((tab1[1,2] + tab1[2,1])/ sum(c(tab1)), 2), "\n")
## 2阶多项式核SVM错判率: 0.4

尝试找到调节参数cost的最优值:

set.seed(101)
res.tune2 <- tune(svm, AHD ~ ., data=d[train,], kernel="polynomial", 
                  order=2, scale=TRUE,
                  ranges=list(cost=c(0.001, 0.01, 0.1, 1, 5, 10, 100, 1000)))
summary(res.tune2)
## 
## Parameter tuning of 'svm':
## 
## - sampling method: 10-fold cross validation 
## 
## - best parameters:
##  cost
##     5
## 
## - best performance: 0.2130952 
## 
## - Detailed performance results:
##    cost     error dispersion
## 1 1e-03 0.4500000 0.08022549
## 2 1e-02 0.4500000 0.08022549
## 3 1e-01 0.4111905 0.09215957
## 4 1e+00 0.2185714 0.09094005
## 5 5e+00 0.2130952 0.09790737
## 6 1e+01 0.2180952 0.07948562
## 7 1e+02 0.2807143 0.09539966
## 8 1e+03 0.2807143 0.09539966
fit.svm2 <- predict(res.tune2$best.model)
tab1 <- table(truth=d[train,"AHD"], fitted=fit.svm2); tab1
##      fitted
## truth  No Yes
##   No  111   3
##   Yes   4  89
cat("2阶多项式核最优参数SVM错判率:", round((tab1[1,2] + tab1[2,1])/ sum(c(tab1)), 2), "\n")
## 2阶多项式核最优参数SVM错判率: 0.03

看这个最优调节参数的模型在测试集上的表现:

pred.svm2 <- predict(res.tune2$best.model, d[test,])
tab1 <- table(truth=d[test,"AHD"], predict=pred.svm2); tab1
##      predict
## truth No Yes
##   No  43   3
##   Yes 10  34
cat("2阶多项式核最优参数SVM测试集错判率:", round((tab1[1,2] + tab1[2,1])/ sum(c(tab1)), 2), "\n")
## 2阶多项式核最优参数SVM测试集错判率: 0.14

在测试集上的表现与线性方法相近。

38.6.1.3 径向核SVM

径向核需要的参数为\(\gamma\)值。 取参数gamma=0.1

res.svm3 <- svm(AHD ~ ., data=d[train,], kernel="radial", 
                gamma=0.1, cost=0.1, scale=TRUE)
fit.svm3 <- predict(res.svm3)
summary(res.svm3)
## 
## Call:
## svm(formula = AHD ~ ., data = d[train, ], kernel = "radial", gamma = 0.1, cost = 0.1, scale = TRUE)
## 
## 
## Parameters:
##    SVM-Type:  C-classification 
##  SVM-Kernel:  radial 
##        cost:  0.1 
## 
## Number of Support Vectors:  179
## 
##  ( 89 90 )
## 
## 
## Number of Classes:  2 
## 
## Levels: 
##  No Yes
tab1 <- table(truth=d[train,"AHD"], fitted=fit.svm3); tab1
##      fitted
## truth  No Yes
##   No  108   6
##   Yes  26  67
cat("径向核(gamma=0.1, cost=0.1)SVM错判率:", round((tab1[1,2] + tab1[2,1])/ sum(c(tab1)), 2), "\n")
## 径向核(gamma=0.1, cost=0.1)SVM错判率: 0.15

选取最优cost, gamma调节参数:

set.seed(101)
res.tune4 <- tune(svm, AHD ~ ., data=d[train,], kernel="radial", 
                  scale=TRUE,
                  ranges=list(cost=c(0.001, 0.01, 0.1, 1, 5, 10, 100, 1000),
                              gamma=c(0.1, 0.01, 0.001)))
summary(res.tune4)
## 
## Parameter tuning of 'svm':
## 
## - sampling method: 10-fold cross validation 
## 
## - best parameters:
##  cost gamma
##   100 0.001
## 
## - best performance: 0.1492857 
## 
## - Detailed performance results:
##     cost gamma     error dispersion
## 1  1e-03 0.100 0.4500000 0.08022549
## 2  1e-02 0.100 0.4500000 0.08022549
## 3  1e-01 0.100 0.2235714 0.09912346
## 4  1e+00 0.100 0.1788095 0.08490543
## 5  5e+00 0.100 0.1835714 0.06267781
## 6  1e+01 0.100 0.1835714 0.07375788
## 7  1e+02 0.100 0.1933333 0.09294732
## 8  1e+03 0.100 0.1933333 0.09294732
## 9  1e-03 0.010 0.4500000 0.08022549
## 10 1e-02 0.010 0.4500000 0.08022549
## 11 1e-01 0.010 0.3147619 0.11998992
## 12 1e+00 0.010 0.1647619 0.06992960
## 13 5e+00 0.010 0.1547619 0.07819776
## 14 1e+01 0.010 0.1547619 0.08135598
## 15 1e+02 0.010 0.2126190 0.06443790
## 16 1e+03 0.010 0.2409524 0.08621108
## 17 1e-03 0.001 0.4500000 0.08022549
## 18 1e-02 0.001 0.4500000 0.08022549
## 19 1e-01 0.001 0.4500000 0.08022549
## 20 1e+00 0.001 0.2138095 0.11215945
## 21 5e+00 0.001 0.1695238 0.07062868
## 22 1e+01 0.001 0.1840476 0.08321647
## 23 1e+02 0.001 0.1492857 0.08228019
## 24 1e+03 0.001 0.1640476 0.07494392
fit.svm4 <- predict(res.tune4$best.model)
tab1 <- table(truth=d[train,"AHD"], fitted=fit.svm4); tab1
##      fitted
## truth  No Yes
##   No  107   7
##   Yes  18  75
cat("径向核最优参数SVM错判率:", round((tab1[1,2] + tab1[2,1])/ sum(c(tab1)), 2), "\n")
## 径向核最优参数SVM错判率: 0.12

看这个最优调节参数的模型在测试集上的表现:

pred.svm4 <- predict(res.tune4$best.model, d[test,])
tab1 <- table(truth=d[test,"AHD"], predict=pred.svm2); tab1
##      predict
## truth No Yes
##   No  43   3
##   Yes 10  34
cat("径向核最优参数SVM测试集错判率:", round((tab1[1,2] + tab1[2,1])/ sum(c(tab1)), 2), "\n")
## 径向核最优参数SVM测试集错判率: 0.14

与线性方法结果相近。

38.7 附录

38.7.1 Hitters数据

knitr::kable(Hitters)
AtBat Hits HmRun Runs RBI Walks Years CAtBat CHits CHmRun CRuns CRBI CWalks League Division PutOuts Assists Errors Salary NewLeague
-Andy Allanson 293 66 1 30 29 14 1 293 66 1 30 29 14 A E 446 33 20 NA A
-Alan Ashby 315 81 7 24 38 39 14 3449 835 69 321 414 375 N W 632 43 10 475.000 N
-Alvin Davis 479 130 18 66 72 76 3 1624 457 63 224 266 263 A W 880 82 14 480.000 A
-Andre Dawson 496 141 20 65 78 37 11 5628 1575 225 828 838 354 N E 200 11 3 500.000 N
-Andres Galarraga 321 87 10 39 42 30 2 396 101 12 48 46 33 N E 805 40 4 91.500 N
-Alfredo Griffin 594 169 4 74 51 35 11 4408 1133 19 501 336 194 A W 282 421 25 750.000 A
-Al Newman 185 37 1 23 8 21 2 214 42 1 30 9 24 N E 76 127 7 70.000 A
-Argenis Salazar 298 73 0 24 24 7 3 509 108 0 41 37 12 A W 121 283 9 100.000 A
-Andres Thomas 323 81 6 26 32 8 2 341 86 6 32 34 8 N W 143 290 19 75.000 N
-Andre Thornton 401 92 17 49 66 65 13 5206 1332 253 784 890 866 A E 0 0 0 1100.000 A
-Alan Trammell 574 159 21 107 75 59 10 4631 1300 90 702 504 488 A E 238 445 22 517.143 A
-Alex Trevino 202 53 4 31 26 27 9 1876 467 15 192 186 161 N W 304 45 11 512.500 N
-Andy VanSlyke 418 113 13 48 61 47 4 1512 392 41 205 204 203 N E 211 11 7 550.000 N
-Alan Wiggins 239 60 0 30 11 22 6 1941 510 4 309 103 207 A E 121 151 6 700.000 A
-Bill Almon 196 43 7 29 27 30 13 3231 825 36 376 290 238 N E 80 45 8 240.000 N
-Billy Beane 183 39 3 20 15 11 3 201 42 3 20 16 11 A W 118 0 0 NA A
-Buddy Bell 568 158 20 89 75 73 15 8068 2273 177 1045 993 732 N W 105 290 10 775.000 N
-Buddy Biancalana 190 46 2 24 8 15 5 479 102 5 65 23 39 A W 102 177 16 175.000 A
-Bruce Bochte 407 104 6 57 43 65 12 5233 1478 100 643 658 653 A W 912 88 9 NA A
-Bruce Bochy 127 32 8 16 22 14 8 727 180 24 67 82 56 N W 202 22 2 135.000 N
-Barry Bonds 413 92 16 72 48 65 1 413 92 16 72 48 65 N E 280 9 5 100.000 N
-Bobby Bonilla 426 109 3 55 43 62 1 426 109 3 55 43 62 A W 361 22 2 115.000 N
-Bob Boone 22 10 1 4 2 1 6 84 26 2 9 9 3 A W 812 84 11 NA A
-Bob Brenly 472 116 16 60 62 74 6 1924 489 67 242 251 240 N W 518 55 3 600.000 N
-Bill Buckner 629 168 18 73 102 40 18 8424 2464 164 1008 1072 402 A E 1067 157 14 776.667 A
-Brett Butler 587 163 4 92 51 70 6 2695 747 17 442 198 317 A E 434 9 3 765.000 A
-Bob Dernier 324 73 4 32 18 22 7 1931 491 13 291 108 180 N E 222 3 3 708.333 N
-Bo Diaz 474 129 10 50 56 40 10 2331 604 61 246 327 166 N W 732 83 13 750.000 N
-Bill Doran 550 152 6 92 37 81 5 2308 633 32 349 182 308 N W 262 329 16 625.000 N
-Brian Downing 513 137 20 90 95 90 14 5201 1382 166 763 734 784 A W 267 5 3 900.000 A
-Bobby Grich 313 84 9 42 30 39 17 6890 1833 224 1033 864 1087 A W 127 221 7 NA A
-Billy Hatcher 419 108 6 55 36 22 3 591 149 8 80 46 31 N W 226 7 4 110.000 N
-Bob Horner 517 141 27 70 87 52 9 3571 994 215 545 652 337 N W 1378 102 8 NA N
-Brook Jacoby 583 168 17 83 80 56 5 1646 452 44 219 208 136 A E 109 292 25 612.500 A
-Bob Kearney 204 49 6 23 25 12 7 1309 308 27 126 132 66 A W 419 46 5 300.000 A
-Bill Madlock 379 106 10 38 60 30 14 6207 1906 146 859 803 571 N W 72 170 24 850.000 N
-Bobby Meacham 161 36 0 19 10 17 4 1053 244 3 156 86 107 A E 70 149 12 NA A
-Bob Melvin 268 60 5 24 25 15 2 350 78 5 34 29 18 N W 442 59 6 90.000 N
-Ben Oglivie 346 98 5 31 53 30 16 5913 1615 235 784 901 560 A E 0 0 0 NA A
-Bip Roberts 241 61 1 34 12 14 1 241 61 1 34 12 14 N W 166 172 10 NA N
-BillyJo Robidoux 181 41 1 15 21 33 2 232 50 4 20 29 45 A E 326 29 5 67.500 A
-Bill Russell 216 54 0 21 18 15 18 7318 1926 46 796 627 483 N W 103 84 5 NA N
-Billy Sample 200 57 6 23 14 14 9 2516 684 46 371 230 195 N W 69 1 1 NA N
-Bill Schroeder 217 46 7 32 19 9 4 694 160 32 86 76 32 A E 307 25 1 180.000 A
-Butch Wynegar 194 40 7 19 29 30 11 4183 1069 64 486 493 608 A E 325 22 2 NA A
-Chris Bando 254 68 2 28 26 22 6 999 236 21 108 117 118 A E 359 30 4 305.000 A
-Chris Brown 416 132 7 57 49 33 3 932 273 24 113 121 80 N W 73 177 18 215.000 N
-Carmen Castillo 205 57 8 34 32 9 5 756 192 32 117 107 51 A E 58 4 4 247.500 A
-Cecil Cooper 542 140 12 46 75 41 16 7099 2130 235 987 1089 431 A E 697 61 9 NA A
-Chili Davis 526 146 13 71 70 84 6 2648 715 77 352 342 289 N W 303 9 9 815.000 N
-Carlton Fisk 457 101 14 42 63 22 17 6521 1767 281 1003 977 619 A W 389 39 4 875.000 A
-Curt Ford 214 53 2 30 29 23 2 226 59 2 32 32 27 N E 109 7 3 70.000 N
-Cliff Johnson 19 7 0 1 2 1 4 41 13 1 3 4 4 A E 0 0 0 NA A
-Carney Lansford 591 168 19 80 72 39 9 4478 1307 113 634 563 319 A W 67 147 4 1200.000 A
-Chet Lemon 403 101 12 45 53 39 12 5150 1429 166 747 666 526 A E 316 6 5 675.000 A
-Candy Maldonado 405 102 18 49 85 20 6 950 231 29 99 138 64 N W 161 10 3 415.000 N
-Carmelo Martinez 244 58 9 28 25 35 4 1335 333 49 164 179 194 N W 142 14 2 340.000 N
-Charlie Moore 235 61 3 24 39 21 14 3926 1029 35 441 401 333 A E 425 43 4 NA A
-Craig Reynolds 313 78 6 32 41 12 12 3742 968 35 409 321 170 N W 106 206 7 416.667 N
-Cal Ripken 627 177 25 98 81 70 6 3210 927 133 529 472 313 A E 240 482 13 1350.000 A
-Cory Snyder 416 113 24 58 69 16 1 416 113 24 58 69 16 A E 203 70 10 90.000 A
-Chris Speier 155 44 6 21 23 15 16 6631 1634 98 698 661 777 N E 53 88 3 275.000 N
-Curt Wilkerson 236 56 0 27 15 11 4 1115 270 1 116 64 57 A W 125 199 13 230.000 A
-Dave Anderson 216 53 1 31 15 22 4 926 210 9 118 69 114 N W 73 152 11 225.000 N
-Doug Baker 24 3 0 1 0 2 3 159 28 0 20 12 9 A W 80 4 0 NA A
-Don Baylor 585 139 31 93 94 62 17 7546 1982 315 1141 1179 727 A E 0 0 0 950.000 A
-Dann Bilardello 191 37 4 12 17 14 4 773 163 16 61 74 52 N E 391 38 8 NA N
-Daryl Boston 199 53 5 29 22 21 3 514 120 8 57 40 39 A W 152 3 5 75.000 A
-Darnell Coles 521 142 20 67 86 45 4 815 205 22 99 103 78 A E 107 242 23 105.000 A
-Dave Collins 419 113 1 44 27 44 12 4484 1231 32 612 344 422 A E 211 2 1 NA A
-Dave Concepcion 311 81 3 42 30 26 17 8247 2198 100 950 909 690 N W 153 223 10 320.000 N
-Darren Daulton 138 31 8 18 21 38 3 244 53 12 33 32 55 N E 244 21 4 NA N
-Doug DeCinces 512 131 26 69 96 52 14 5347 1397 221 712 815 548 A W 119 216 12 850.000 A
-Darrell Evans 507 122 29 78 85 91 18 7761 1947 347 1175 1152 1380 A E 808 108 2 535.000 A
-Dwight Evans 529 137 26 86 97 97 15 6661 1785 291 1082 949 989 A E 280 10 5 933.333 A
-Damaso Garcia 424 119 6 57 46 13 9 3651 1046 32 461 301 112 A E 224 286 8 850.000 N
-Dan Gladden 351 97 4 55 29 39 4 1258 353 16 196 110 117 N W 226 7 3 210.000 A
-Danny Heep 195 55 5 24 33 30 8 1313 338 25 144 149 153 N E 83 2 1 NA N
-Dave Henderson 388 103 15 59 47 39 6 2174 555 80 285 274 186 A W 182 9 4 325.000 A
-Donnie Hill 339 96 4 37 29 23 4 1064 290 11 123 108 55 A W 104 213 9 275.000 A
-Dave Kingman 561 118 35 70 94 33 16 6677 1575 442 901 1210 608 A W 463 32 8 NA A
-Davey Lopes 255 70 7 49 35 43 15 6311 1661 154 1019 608 820 N E 51 54 8 450.000 N
-Don Mattingly 677 238 31 117 113 53 5 2223 737 93 349 401 171 A E 1377 100 6 1975.000 A
-Darryl Motley 227 46 7 23 20 12 5 1325 324 44 156 158 67 A W 92 2 2 NA A
-Dale Murphy 614 163 29 89 83 75 11 5017 1388 266 813 822 617 N W 303 6 6 1900.000 N
-Dwayne Murphy 329 83 9 50 39 56 9 3828 948 145 575 528 635 A W 276 6 2 600.000 A
-Dave Parker 637 174 31 89 116 56 14 6727 2024 247 978 1093 495 N W 278 9 9 1041.667 N
-Dan Pasqua 280 82 16 44 45 47 2 428 113 25 61 70 63 A E 148 4 2 110.000 A
-Darrell Porter 155 41 12 21 29 22 16 5409 1338 181 746 805 875 A W 165 9 1 260.000 A
-Dick Schofield 458 114 13 67 57 48 4 1350 298 28 160 123 122 A W 246 389 18 475.000 A
-Don Slaught 314 83 13 39 46 16 5 1457 405 28 156 159 76 A W 533 40 4 431.500 A
-Darryl Strawberry 475 123 27 76 93 72 4 1810 471 108 292 343 267 N E 226 10 6 1220.000 N
-Dale Sveum 317 78 7 35 35 32 1 317 78 7 35 35 32 A E 45 122 26 70.000 A
-Danny Tartabull 511 138 25 76 96 61 3 592 164 28 87 110 71 A W 157 7 8 145.000 A
-Dickie Thon 278 69 3 24 21 29 8 2079 565 32 258 192 162 N W 142 210 10 NA N
-Denny Walling 382 119 13 54 58 36 12 2133 594 41 287 294 227 N W 59 156 9 595.000 N
-Dave Winfield 565 148 24 90 104 77 14 7287 2083 305 1135 1234 791 A E 292 9 5 1861.460 A
-Enos Cabell 277 71 2 27 29 14 15 5952 1647 60 753 596 259 N W 360 32 5 NA N
-Eric Davis 415 115 27 97 71 68 3 711 184 45 156 119 99 N W 274 2 7 300.000 N
-Eddie Milner 424 110 15 70 47 36 7 2130 544 38 335 174 258 N W 292 6 3 490.000 N
-Eddie Murray 495 151 17 61 84 78 10 5624 1679 275 884 1015 709 A E 1045 88 13 2460.000 A
-Ernest Riles 524 132 9 69 47 54 2 972 260 14 123 92 90 A E 212 327 20 NA A
-Ed Romero 233 49 2 41 23 18 8 1350 336 7 166 122 106 A E 102 132 10 375.000 A
-Ernie Whitt 395 106 16 48 56 35 10 2303 571 86 266 323 248 A E 709 41 7 NA A
-Fred Lynn 397 114 23 67 67 53 13 5589 1632 241 906 926 716 A E 244 2 4 NA A
-Floyd Rayford 210 37 8 15 19 15 6 994 244 36 107 114 53 A E 40 115 15 NA A
-Franklin Stubbs 420 95 23 55 58 37 3 646 139 31 77 77 61 N W 206 10 7 NA N
-Frank White 566 154 22 76 84 43 14 6100 1583 131 743 693 300 A W 316 439 10 750.000 A
-George Bell 641 198 31 101 108 41 5 2129 610 92 297 319 117 A E 269 17 10 1175.000 A
-Glenn Braggs 215 51 4 19 18 11 1 215 51 4 19 18 11 A E 116 5 12 70.000 A
-George Brett 441 128 16 70 73 80 14 6675 2095 209 1072 1050 695 A W 97 218 16 1500.000 A
-Greg Brock 325 76 16 33 52 37 5 1506 351 71 195 219 214 N W 726 87 3 385.000 A
-Gary Carter 490 125 24 81 105 62 13 6063 1646 271 847 999 680 N E 869 62 8 1925.571 N
-Glenn Davis 574 152 31 91 101 64 3 985 260 53 148 173 95 N W 1253 111 11 215.000 N
-George Foster 284 64 14 30 42 24 18 7023 1925 348 986 1239 666 N E 96 4 4 NA N
-Gary Gaetti 596 171 34 91 108 52 6 2862 728 107 361 401 224 A W 118 334 21 900.000 A
-Greg Gagne 472 118 12 63 54 30 4 793 187 14 102 80 50 A W 228 377 26 155.000 A
-George Hendrick 283 77 14 45 47 26 16 6840 1910 259 915 1067 546 A W 144 6 5 700.000 A
-Glenn Hubbard 408 94 4 42 36 66 9 3573 866 59 429 365 410 N W 282 487 19 535.000 N
-Garth Iorg 327 85 3 30 44 20 8 2140 568 16 216 208 93 A E 91 185 12 362.500 A
-Gary Matthews 370 96 21 49 46 60 15 6986 1972 231 1070 955 921 N E 137 5 9 733.333 N
-Graig Nettles 354 77 16 36 55 41 20 8716 2172 384 1172 1267 1057 N W 83 174 16 200.000 N
-Gary Pettis 539 139 5 93 58 69 5 1469 369 12 247 126 198 A W 462 9 7 400.000 A
-Gary Redus 340 84 11 62 33 47 5 1516 376 42 284 141 219 N E 185 8 4 400.000 A
-Garry Templeton 510 126 2 42 44 35 11 5562 1578 44 703 519 256 N W 207 358 20 737.500 N
-Gorman Thomas 315 59 16 45 36 58 13 4677 1051 268 681 782 697 A W 0 0 0 NA A
-Greg Walker 282 78 13 37 51 29 5 1649 453 73 211 280 138 A W 670 57 5 500.000 A
-Gary Ward 380 120 5 54 51 31 8 3118 900 92 444 419 240 A W 237 8 1 600.000 A
-Glenn Wilson 584 158 15 70 84 42 5 2358 636 58 265 316 134 N E 331 20 4 662.500 N
-Harold Baines 570 169 21 72 88 38 7 3754 1077 140 492 589 263 A W 295 15 5 950.000 A
-Hubie Brooks 306 104 14 50 58 25 7 2954 822 55 313 377 187 N E 116 222 15 750.000 N
-Howard Johnson 220 54 10 30 39 31 5 1185 299 40 145 154 128 N E 50 136 20 297.500 N
-Hal McRae 278 70 7 22 37 18 18 7186 2081 190 935 1088 643 A W 0 0 0 325.000 A
-Harold Reynolds 445 99 1 46 24 29 4 618 129 1 72 31 48 A W 278 415 16 87.500 A
-Harry Spilman 143 39 5 18 30 15 9 639 151 16 80 97 61 N W 138 15 1 175.000 N
-Herm Winningham 185 40 4 23 11 18 3 524 125 7 58 37 47 N E 97 2 2 90.000 N
-Jesse Barfield 589 170 40 107 108 69 6 2325 634 128 371 376 238 A E 368 20 3 1237.500 A
-Juan Beniquez 343 103 6 48 36 40 15 4338 1193 70 581 421 325 A E 211 56 13 430.000 A
-Juan Bonilla 284 69 1 33 18 25 5 1407 361 6 139 98 111 A E 122 140 5 NA N
-John Cangelosi 438 103 2 65 32 71 2 440 103 2 67 32 71 A W 276 7 9 100.000 N
-Jose Canseco 600 144 33 85 117 65 2 696 173 38 101 130 69 A W 319 4 14 165.000 A
-Joe Carter 663 200 29 108 121 32 4 1447 404 57 210 222 68 A E 241 8 6 250.000 A
-Jack Clark 232 55 9 34 23 45 12 4405 1213 194 702 705 625 N E 623 35 3 1300.000 N
-Jose Cruz 479 133 10 48 72 55 17 7472 2147 153 980 1032 854 N W 237 5 4 773.333 N
-Julio Cruz 209 45 0 38 19 42 10 3859 916 23 557 279 478 A W 132 205 5 NA A
-Jody Davis 528 132 21 61 74 41 6 2641 671 97 273 383 226 N E 885 105 8 1008.333 N
-Jim Dwyer 160 39 8 18 31 22 14 2128 543 56 304 268 298 A E 33 3 0 275.000 A
-Julio Franco 599 183 10 80 74 32 5 2482 715 27 330 326 158 A E 231 374 18 775.000 A
-Jim Gantner 497 136 7 58 38 26 11 3871 1066 40 450 367 241 A E 304 347 10 850.000 A
-Johnny Grubb 210 70 13 32 51 28 15 4040 1130 97 544 462 551 A E 0 0 0 365.000 A
-Jerry Hairston 225 61 5 32 26 26 11 1568 408 25 202 185 257 A W 132 9 0 NA A
-Jack Howell 151 41 4 26 21 19 2 288 68 9 45 39 35 A W 28 56 2 95.000 A
-John Kruk 278 86 4 33 38 45 1 278 86 4 33 38 45 N W 102 4 2 110.000 N
-Jeffrey Leonard 341 95 6 48 42 20 10 2964 808 81 379 428 221 N W 158 4 5 100.000 N
-Jim Morrison 537 147 23 58 88 47 10 2744 730 97 302 351 174 N E 92 257 20 277.500 N
-John Moses 399 102 3 56 34 34 5 670 167 4 89 48 54 A W 211 9 3 80.000 A
-Jerry Mumphrey 309 94 5 37 32 26 13 4618 1330 57 616 522 436 N E 161 3 3 600.000 N
-Joe Orsulak 401 100 2 60 19 28 4 876 238 2 126 44 55 N E 193 11 4 NA N
-Jorge Orta 336 93 9 35 46 23 15 5779 1610 128 730 741 497 A W 0 0 0 NA A
-Jim Presley 616 163 27 83 107 32 3 1437 377 65 181 227 82 A W 110 308 15 200.000 A
-Jamie Quirk 219 47 8 24 26 17 12 1188 286 23 100 125 63 A W 260 58 4 NA A
-Johnny Ray 579 174 7 67 78 58 6 3053 880 32 366 337 218 N E 280 479 5 657.000 N
-Jeff Reed 165 39 2 13 9 16 3 196 44 2 18 10 18 A W 332 19 2 75.000 N
-Jim Rice 618 200 20 98 110 62 13 7127 2163 351 1104 1289 564 A E 330 16 8 2412.500 A
-Jerry Royster 257 66 5 31 26 32 14 3910 979 33 518 324 382 N W 87 166 14 250.000 A
-John Russell 315 76 13 35 60 25 3 630 151 24 68 94 55 N E 498 39 13 155.000 N
-Juan Samuel 591 157 16 90 78 26 4 2020 541 52 310 226 91 N E 290 440 25 640.000 N
-John Shelby 404 92 11 54 49 18 6 1354 325 30 188 135 63 A E 222 5 5 300.000 A
-Joel Skinner 315 73 5 23 37 16 4 450 108 6 38 46 28 A W 227 15 3 110.000 A
-Jeff Stone 249 69 6 32 19 20 4 702 209 10 97 48 44 N E 103 8 2 NA N
-Jim Sundberg 429 91 12 41 42 57 13 5590 1397 83 578 579 644 A W 686 46 4 825.000 N
-Jim Traber 212 54 13 28 44 18 2 233 59 13 31 46 20 A E 243 23 5 NA A
-Jose Uribe 453 101 3 46 43 61 3 948 218 6 96 72 91 N W 249 444 16 195.000 N
-Jerry Willard 161 43 4 17 26 22 3 707 179 21 77 99 76 A W 300 12 2 NA A
-Joel Youngblood 184 47 5 20 28 18 11 3327 890 74 419 382 304 N W 49 2 0 450.000 N
-Kevin Bass 591 184 20 83 79 38 5 1689 462 40 219 195 82 N W 303 12 5 630.000 N
-Kal Daniels 181 58 6 34 23 22 1 181 58 6 34 23 22 N W 88 0 3 86.500 N
-Kirk Gibson 441 118 28 84 86 68 8 2723 750 126 433 420 309 A E 190 2 2 1300.000 A
-Ken Griffey 490 150 21 69 58 35 14 6126 1839 121 983 707 600 A E 96 5 3 1000.000 N
-Keith Hernandez 551 171 13 94 83 94 13 6090 1840 128 969 900 917 N E 1199 149 5 1800.000 N
-Kent Hrbek 550 147 29 85 91 71 6 2816 815 117 405 474 319 A W 1218 104 10 1310.000 A
-Ken Landreaux 283 74 4 34 29 22 10 3919 1062 85 505 456 283 N W 145 5 7 737.500 N
-Kevin McReynolds 560 161 26 89 96 66 4 1789 470 65 233 260 155 N W 332 9 8 625.000 N
-Kevin Mitchell 328 91 12 51 43 33 2 342 94 12 51 44 33 N E 145 59 8 125.000 N
-Keith Moreland 586 159 12 72 79 53 9 3082 880 83 363 477 295 N E 181 13 4 1043.333 N
-Ken Oberkfell 503 136 5 62 48 83 10 3423 970 20 408 303 414 N W 65 258 8 725.000 N
-Ken Phelps 344 85 24 69 64 88 7 911 214 64 150 156 187 A W 0 0 0 300.000 A
-Kirby Puckett 680 223 31 119 96 34 3 1928 587 35 262 201 91 A W 429 8 6 365.000 A
-Kurt Stillwell 279 64 0 31 26 30 1 279 64 0 31 26 30 N W 107 205 16 75.000 N
-Leon Durham 484 127 20 66 65 67 7 3006 844 116 436 458 377 N E 1231 80 7 1183.333 N
-Len Dykstra 431 127 8 77 45 58 2 667 187 9 117 64 88 N E 283 8 3 202.500 N
-Larry Herndon 283 70 8 33 37 27 12 4479 1222 94 557 483 307 A E 156 2 2 225.000 A
-Lee Lacy 491 141 11 77 47 37 15 4291 1240 84 615 430 340 A E 239 8 2 525.000 A
-Len Matuszek 199 52 9 26 28 21 6 805 191 30 113 119 87 N W 235 22 5 265.000 N
-Lloyd Moseby 589 149 21 89 86 64 7 3558 928 102 513 471 351 A E 371 6 6 787.500 A
-Lance Parrish 327 84 22 53 62 38 10 4273 1123 212 577 700 334 A E 483 48 6 800.000 N
-Larry Parrish 464 128 28 67 94 52 13 5829 1552 210 740 840 452 A W 0 0 0 587.500 A
-Luis Rivera 166 34 0 20 13 17 1 166 34 0 20 13 17 N E 64 119 9 NA N
-Larry Sheets 338 92 18 42 60 21 3 682 185 36 88 112 50 A E 0 0 0 145.000 A
-Lonnie Smith 508 146 8 80 44 46 9 3148 915 41 571 289 326 A W 245 5 9 NA A
-Lou Whitaker 584 157 20 95 73 63 10 4704 1320 93 724 522 576 A E 276 421 11 420.000 A
-Mike Aldrete 216 54 2 27 25 33 1 216 54 2 27 25 33 N W 317 36 1 75.000 N
-Marty Barrett 625 179 4 94 60 65 5 1696 476 12 216 163 166 A E 303 450 14 575.000 A
-Mike Brown 243 53 4 18 26 27 4 853 228 23 101 110 76 N E 107 3 3 NA N
-Mike Davis 489 131 19 77 55 34 7 2051 549 62 300 263 153 A W 310 9 9 780.000 A
-Mike Diaz 209 56 12 22 36 19 2 216 58 12 24 37 19 N E 201 6 3 90.000 N
-Mariano Duncan 407 93 8 47 30 30 2 969 230 14 121 69 68 N W 172 317 25 150.000 N
-Mike Easler 490 148 14 64 78 49 13 3400 1000 113 445 491 301 A E 0 0 0 700.000 N
-Mike Fitzgerald 209 59 6 20 37 27 4 884 209 14 66 106 92 N E 415 35 3 NA N
-Mel Hall 442 131 18 68 77 33 6 1416 398 47 210 203 136 A E 233 7 7 550.000 A
-Mickey Hatcher 317 88 3 40 32 19 8 2543 715 28 269 270 118 A W 220 16 4 NA A
-Mike Heath 288 65 8 30 36 27 9 2815 698 55 315 325 189 N E 259 30 10 650.000 A
-Mike Kingery 209 54 3 25 14 12 1 209 54 3 25 14 12 A W 102 6 3 68.000 A
-Mike LaValliere 303 71 3 18 30 36 3 344 76 3 20 36 45 N E 468 47 6 100.000 N
-Mike Marshall 330 77 19 47 53 27 6 1928 516 90 247 288 161 N W 149 8 6 670.000 N
-Mike Pagliarulo 504 120 28 71 71 54 3 1085 259 54 150 167 114 A E 103 283 19 175.000 A
-Mark Salas 258 60 8 28 33 18 3 638 170 17 80 75 36 A W 358 32 8 137.000 A
-Mike Schmidt 20 1 0 0 0 0 2 41 9 2 6 7 4 N E 78 220 6 2127.333 N
-Mike Scioscia 374 94 5 36 26 62 7 1968 519 26 181 199 288 N W 756 64 15 875.000 N
-Mickey Tettleton 211 43 10 26 35 39 3 498 116 14 59 55 78 A W 463 32 8 120.000 A
-Milt Thompson 299 75 6 38 23 26 3 580 160 8 71 33 44 N E 212 1 2 140.000 N
-Mitch Webster 576 167 8 89 49 57 4 822 232 19 132 83 79 N E 325 12 8 210.000 N
-Mookie Wilson 381 110 9 61 45 32 7 3015 834 40 451 249 168 N E 228 7 5 800.000 N
-Marvell Wynne 288 76 7 34 37 15 4 1644 408 16 198 120 113 N W 203 3 3 240.000 N
-Mike Young 369 93 9 43 42 49 5 1258 323 54 181 177 157 A E 149 1 6 350.000 A
-Nick Esasky 330 76 12 35 41 47 4 1367 326 55 167 198 167 N W 512 30 5 NA N
-Ozzie Guillen 547 137 2 58 47 12 2 1038 271 3 129 80 24 A W 261 459 22 175.000 A
-Oddibe McDowell 572 152 18 105 49 65 2 978 249 36 168 91 101 A W 325 13 3 200.000 A
-Omar Moreno 359 84 4 46 27 21 12 4992 1257 37 699 386 387 N W 151 8 5 NA N
-Ozzie Smith 514 144 0 67 54 79 9 4739 1169 13 583 374 528 N E 229 453 15 1940.000 N
-Ozzie Virgil 359 80 15 45 48 63 7 1493 359 61 176 202 175 N W 682 93 13 700.000 N
-Phil Bradley 526 163 12 88 50 77 4 1556 470 38 245 167 174 A W 250 11 1 750.000 A
-Phil Garner 313 83 9 43 41 30 14 5885 1543 104 751 714 535 N W 58 141 23 450.000 N
-Pete Incaviglia 540 135 30 82 88 55 1 540 135 30 82 88 55 A W 157 6 14 172.000 A
-Paul Molitor 437 123 9 62 55 40 9 4139 1203 79 676 390 364 A E 82 170 15 1260.000 A
-Pete O’Brien 551 160 23 86 90 87 5 2235 602 75 278 328 273 A W 1224 115 11 NA A
-Pete Rose 237 52 0 15 25 30 24 14053 4256 160 2165 1314 1566 N W 523 43 6 750.000 N
-Pat Sheridan 236 56 6 41 19 21 5 1257 329 24 166 125 105 A E 172 1 4 190.000 A
-Pat Tabler 473 154 6 61 48 29 6 1966 566 29 250 252 178 A E 846 84 9 580.000 A
-Rafael Belliard 309 72 0 33 31 26 5 354 82 0 41 32 26 N E 117 269 12 130.000 N
-Rick Burleson 271 77 5 35 29 33 12 4933 1358 48 630 435 403 A W 62 90 3 450.000 A
-Randy Bush 357 96 7 50 45 39 5 1394 344 43 178 192 136 A W 167 2 4 300.000 A
-Rick Cerone 216 56 4 22 18 15 12 2796 665 43 266 304 198 A E 391 44 4 250.000 A
-Ron Cey 256 70 13 42 36 44 16 7058 1845 312 965 1128 990 N E 41 118 8 1050.000 A
-Rob Deer 466 108 33 75 86 72 3 652 142 44 102 109 102 A E 286 8 8 215.000 A
-Rick Dempsey 327 68 13 42 29 45 18 3949 939 78 438 380 466 A E 659 53 7 400.000 A
-Rich Gedman 462 119 16 49 65 37 7 2131 583 69 244 288 150 A E 866 65 6 NA A
-Ron Hassey 341 110 9 45 49 46 9 2331 658 50 249 322 274 A E 251 9 4 560.000 A
-Rickey Henderson 608 160 28 130 74 89 8 4071 1182 103 862 417 708 A E 426 4 6 1670.000 A
-Reggie Jackson 419 101 18 65 58 92 20 9528 2510 548 1509 1659 1342 A W 0 0 0 487.500 A
-Ricky Jones 33 6 0 2 4 7 1 33 6 0 2 4 7 A W 205 5 4 NA A
-Ron Kittle 376 82 21 42 60 35 5 1770 408 115 238 299 157 A W 0 0 0 425.000 A
-Ray Knight 486 145 11 51 76 40 11 3967 1102 67 410 497 284 N E 88 204 16 500.000 A
-Randy Kutcher 186 44 7 28 16 11 1 186 44 7 28 16 11 N W 99 3 1 NA N
-Rudy Law 307 80 1 42 36 29 7 2421 656 18 379 198 184 A W 145 2 2 NA A
-Rick Leach 246 76 5 35 39 13 6 912 234 12 102 96 80 A E 44 0 1 250.000 A
-Rick Manning 205 52 8 31 27 17 12 5134 1323 56 643 445 459 A E 155 3 2 400.000 A
-Rance Mulliniks 348 90 11 50 45 43 10 2288 614 43 295 273 269 A E 60 176 6 450.000 A
-Ron Oester 523 135 8 52 44 52 9 3368 895 39 377 284 296 N W 367 475 19 750.000 N
-Rey Quinones 312 68 2 32 22 24 1 312 68 2 32 22 24 A E 86 150 15 70.000 A
-Rafael Ramirez 496 119 8 57 33 21 7 3358 882 36 365 280 165 N W 155 371 29 875.000 N
-Ronn Reynolds 126 27 3 8 10 5 4 239 49 3 16 13 14 N E 190 2 9 190.000 N
-Ron Roenicke 275 68 5 42 42 61 6 961 238 16 128 104 172 N E 181 3 2 191.000 N
-Ryne Sandberg 627 178 14 68 76 46 6 3146 902 74 494 345 242 N E 309 492 5 740.000 N
-Rafael Santana 394 86 1 38 28 36 4 1089 267 3 94 71 76 N E 203 369 16 250.000 N
-Rick Schu 208 57 8 32 25 18 3 653 170 17 98 54 62 N E 42 94 13 140.000 N
-Ruben Sierra 382 101 16 50 55 22 1 382 101 16 50 55 22 A W 200 7 6 97.500 A
-Roy Smalley 459 113 20 59 57 68 12 5348 1369 155 713 660 735 A W 0 0 0 740.000 A
-Robby Thompson 549 149 7 73 47 42 1 549 149 7 73 47 42 N W 255 450 17 140.000 N
-Rob Wilfong 288 63 3 25 33 16 10 2682 667 38 315 259 204 A W 135 257 7 341.667 A
-Reggie Williams 303 84 4 35 32 23 2 312 87 4 39 32 23 N W 179 5 3 NA N
-Robin Yount 522 163 9 82 46 62 13 7037 2019 153 1043 827 535 A E 352 9 1 1000.000 A
-Steve Balboni 512 117 29 54 88 43 6 1750 412 100 204 276 155 A W 1236 98 18 100.000 A
-Scott Bradley 220 66 5 20 28 13 3 290 80 5 27 31 15 A W 281 21 3 90.000 A
-Sid Bream 522 140 16 73 77 60 4 730 185 22 93 106 86 N E 1320 166 17 200.000 N
-Steve Buechele 461 112 18 54 54 35 2 680 160 24 76 75 49 A W 111 226 11 135.000 A
-Shawon Dunston 581 145 17 66 68 21 2 831 210 21 106 86 40 N E 320 465 32 155.000 N
-Scott Fletcher 530 159 3 82 50 47 6 1619 426 11 218 149 163 A W 196 354 15 475.000 A
-Steve Garvey 557 142 21 58 81 23 18 8759 2583 271 1138 1299 478 N W 1160 53 7 1450.000 N
-Steve Jeltz 439 96 0 44 36 65 4 711 148 1 68 56 99 N E 229 406 22 150.000 N
-Steve Lombardozzi 453 103 8 53 33 52 2 507 123 8 63 39 58 A W 289 407 6 105.000 A
-Spike Owen 528 122 1 67 45 51 4 1716 403 12 211 146 155 A W 209 372 17 350.000 A
-Steve Sax 633 210 6 91 56 59 6 3070 872 19 420 230 274 N W 367 432 16 90.000 N
-Tony Armas 16 2 0 1 0 0 2 28 4 0 1 0 0 A E 247 4 8 NA A
-Tony Bernazard 562 169 17 88 73 53 8 3181 841 61 450 342 373 A E 351 442 17 530.000 A
-Tom Brookens 281 76 3 42 25 20 8 2658 657 48 324 300 179 A E 106 144 7 341.667 A
-Tom Brunansky 593 152 23 69 75 53 6 2765 686 133 369 384 321 A W 315 10 6 940.000 A
-Tony Fernandez 687 213 10 91 65 27 4 1518 448 15 196 137 89 A E 294 445 13 350.000 A
-Tim Flannery 368 103 3 48 28 54 8 1897 493 9 207 162 198 N W 209 246 3 326.667 N
-Tom Foley 263 70 1 26 23 30 4 888 220 9 83 82 86 N E 81 147 4 250.000 N
-Tony Gwynn 642 211 14 107 59 52 5 2364 770 27 352 230 193 N W 337 19 4 740.000 N
-Terry Harper 265 68 8 26 30 29 7 1337 339 32 135 163 128 N W 92 5 3 425.000 A
-Toby Harrah 289 63 7 36 41 44 17 7402 1954 195 1115 919 1153 A W 166 211 7 NA A
-Tommy Herr 559 141 2 48 61 73 8 3162 874 16 421 349 359 N E 352 414 9 925.000 N
-Tim Hulett 520 120 17 53 44 21 4 927 227 22 106 80 52 A W 70 144 11 185.000 A
-Terry Kennedy 19 4 1 2 3 1 1 19 4 1 2 3 1 N W 692 70 8 920.000 A
-Tito Landrum 205 43 2 24 17 20 7 854 219 12 105 99 71 N E 131 6 1 286.667 N
-Tim Laudner 193 47 10 21 29 24 6 1136 256 42 129 139 106 A W 299 13 5 245.000 A
-Tom O’Malley 181 46 1 19 18 17 5 937 238 9 88 95 104 A E 37 98 9 NA A
-Tom Paciorek 213 61 4 17 22 3 17 4061 1145 83 488 491 244 A W 178 45 4 235.000 A
-Tony Pena 510 147 10 56 52 53 7 2872 821 63 307 340 174 N E 810 99 18 1150.000 N
-Terry Pendleton 578 138 1 56 59 34 3 1399 357 7 149 161 87 N E 133 371 20 160.000 N
-Tony Perez 200 51 2 14 29 25 23 9778 2732 379 1272 1652 925 N W 398 29 7 NA N
-Tony Phillips 441 113 5 76 52 76 5 1546 397 17 226 149 191 A W 160 290 11 425.000 A
-Terry Puhl 172 42 3 17 14 15 10 4086 1150 57 579 363 406 N W 65 0 0 900.000 N
-Tim Raines 580 194 9 91 62 78 8 3372 1028 48 604 314 469 N E 270 13 6 NA N
-Ted Simmons 127 32 4 14 25 12 19 8396 2402 242 1048 1348 819 N W 167 18 6 500.000 N
-Tim Teufel 279 69 4 35 31 32 4 1359 355 31 180 148 158 N E 133 173 9 277.500 N
-Tim Wallach 480 112 18 50 71 44 7 3031 771 110 338 406 239 N E 94 270 16 750.000 N
-Vince Coleman 600 139 0 94 29 60 2 1236 309 1 201 69 110 N E 300 12 9 160.000 N
-Von Hayes 610 186 19 107 98 74 6 2728 753 69 399 366 286 N E 1182 96 13 1300.000 N
-Vance Law 360 81 5 37 44 37 7 2268 566 41 279 257 246 N E 170 284 3 525.000 N
-Wally Backman 387 124 1 67 27 36 7 1775 506 6 272 125 194 N E 186 290 17 550.000 N
-Wade Boggs 580 207 8 107 71 105 5 2778 978 32 474 322 417 A E 121 267 19 1600.000 A
-Will Clark 408 117 11 66 41 34 1 408 117 11 66 41 34 N W 942 72 11 120.000 N
-Wally Joyner 593 172 22 82 100 57 1 593 172 22 82 100 57 A W 1222 139 15 165.000 A
-Wayne Krenchicki 221 53 2 21 23 22 8 1063 283 15 107 124 106 N E 325 58 6 NA N
-Willie McGee 497 127 7 65 48 37 5 2703 806 32 379 311 138 N E 325 9 3 700.000 N
-Willie Randolph 492 136 5 76 50 94 12 5511 1511 39 897 451 875 A E 313 381 20 875.000 A
-Wayne Tolleson 475 126 3 61 43 52 6 1700 433 7 217 93 146 A W 37 113 7 385.000 A
-Willie Upshaw 573 144 9 85 60 78 8 3198 857 97 470 420 332 A E 1314 131 12 960.000 A
-Willie Wilson 631 170 9 77 44 31 11 4908 1457 30 775 357 249 A W 408 4 3 1000.000 A

38.7.2 Heart数据

knitr::kable(Heart)
Age Sex ChestPain RestBP Chol Fbs RestECG MaxHR ExAng Oldpeak Slope Ca Thal AHD
63 1 typical 145 233 1 2 150 0 2.3 3 0 fixed No
67 1 asymptomatic 160 286 0 2 108 1 1.5 2 3 normal Yes
67 1 asymptomatic 120 229 0 2 129 1 2.6 2 2 reversable Yes
37 1 nonanginal 130 250 0 0 187 0 3.5 3 0 normal No
41 0 nontypical 130 204 0 2 172 0 1.4 1 0 normal No
56 1 nontypical 120 236 0 0 178 0 0.8 1 0 normal No
62 0 asymptomatic 140 268 0 2 160 0 3.6 3 2 normal Yes
57 0 asymptomatic 120 354 0 0 163 1 0.6 1 0 normal No
63 1 asymptomatic 130 254 0 2 147 0 1.4 2 1 reversable Yes
53 1 asymptomatic 140 203 1 2 155 1 3.1 3 0 reversable Yes
57 1 asymptomatic 140 192 0 0 148 0 0.4 2 0 fixed No
56 0 nontypical 140 294 0 2 153 0 1.3 2 0 normal No
56 1 nonanginal 130 256 1 2 142 1 0.6 2 1 fixed Yes
44 1 nontypical 120 263 0 0 173 0 0.0 1 0 reversable No
52 1 nonanginal 172 199 1 0 162 0 0.5 1 0 reversable No
57 1 nonanginal 150 168 0 0 174 0 1.6 1 0 normal No
48 1 nontypical 110 229 0 0 168 0 1.0 3 0 reversable Yes
54 1 asymptomatic 140 239 0 0 160 0 1.2 1 0 normal No
48 0 nonanginal 130 275 0 0 139 0 0.2 1 0 normal No
49 1 nontypical 130 266 0 0 171 0 0.6 1 0 normal No
64 1 typical 110 211 0 2 144 1 1.8 2 0 normal No
58 0 typical 150 283 1 2 162 0 1.0 1 0 normal No
58 1 nontypical 120 284 0 2 160 0 1.8 2 0 normal Yes
58 1 nonanginal 132 224 0 2 173 0 3.2 1 2 reversable Yes
60 1 asymptomatic 130 206 0 2 132 1 2.4 2 2 reversable Yes
50 0 nonanginal 120 219 0 0 158 0 1.6 2 0 normal No
58 0 nonanginal 120 340 0 0 172 0 0.0 1 0 normal No
66 0 typical 150 226 0 0 114 0 2.6 3 0 normal No
43 1 asymptomatic 150 247 0 0 171 0 1.5 1 0 normal No
40 1 asymptomatic 110 167 0 2 114 1 2.0 2 0 reversable Yes
69 0 typical 140 239 0 0 151 0 1.8 1 2 normal No
60 1 asymptomatic 117 230 1 0 160 1 1.4 1 2 reversable Yes
64 1 nonanginal 140 335 0 0 158 0 0.0 1 0 normal Yes
59 1 asymptomatic 135 234 0 0 161 0 0.5 2 0 reversable No
44 1 nonanginal 130 233 0 0 179 1 0.4 1 0 normal No
42 1 asymptomatic 140 226 0 0 178 0 0.0 1 0 normal No
43 1 asymptomatic 120 177 0 2 120 1 2.5 2 0 reversable Yes
57 1 asymptomatic 150 276 0 2 112 1 0.6 2 1 fixed Yes
55 1 asymptomatic 132 353 0 0 132 1 1.2 2 1 reversable Yes
61 1 nonanginal 150 243 1 0 137 1 1.0 2 0 normal No
65 0 asymptomatic 150 225 0 2 114 0 1.0 2 3 reversable Yes
40 1 typical 140 199 0 0 178 1 1.4 1 0 reversable No
71 0 nontypical 160 302 0 0 162 0 0.4 1 2 normal No
59 1 nonanginal 150 212 1 0 157 0 1.6 1 0 normal No
61 0 asymptomatic 130 330 0 2 169 0 0.0 1 0 normal Yes
58 1 nonanginal 112 230 0 2 165 0 2.5 2 1 reversable Yes
51 1 nonanginal 110 175 0 0 123 0 0.6 1 0 normal No
50 1 asymptomatic 150 243 0 2 128 0 2.6 2 0 reversable Yes
65 0 nonanginal 140 417 1 2 157 0 0.8 1 1 normal No
53 1 nonanginal 130 197 1 2 152 0 1.2 3 0 normal No
41 0 nontypical 105 198 0 0 168 0 0.0 1 1 normal No
65 1 asymptomatic 120 177 0 0 140 0 0.4 1 0 reversable No
44 1 asymptomatic 112 290 0 2 153 0 0.0 1 1 normal Yes
44 1 nontypical 130 219 0 2 188 0 0.0 1 0 normal No
60 1 asymptomatic 130 253 0 0 144 1 1.4 1 1 reversable Yes
54 1 asymptomatic 124 266 0 2 109 1 2.2 2 1 reversable Yes
50 1 nonanginal 140 233 0 0 163 0 0.6 2 1 reversable Yes
41 1 asymptomatic 110 172 0 2 158 0 0.0 1 0 reversable Yes
54 1 nonanginal 125 273 0 2 152 0 0.5 3 1 normal No
51 1 typical 125 213 0 2 125 1 1.4 1 1 normal No
51 0 asymptomatic 130 305 0 0 142 1 1.2 2 0 reversable Yes
46 0 nonanginal 142 177 0 2 160 1 1.4 3 0 normal No
58 1 asymptomatic 128 216 0 2 131 1 2.2 2 3 reversable Yes
54 0 nonanginal 135 304 1 0 170 0 0.0 1 0 normal No
54 1 asymptomatic 120 188 0 0 113 0 1.4 2 1 reversable Yes
60 1 asymptomatic 145 282 0 2 142 1 2.8 2 2 reversable Yes
60 1 nonanginal 140 185 0 2 155 0 3.0 2 0 normal Yes
54 1 nonanginal 150 232 0 2 165 0 1.6 1 0 reversable No
59 1 asymptomatic 170 326 0 2 140 1 3.4 3 0 reversable Yes
46 1 nonanginal 150 231 0 0 147 0 3.6 2 0 normal Yes
65 0 nonanginal 155 269 0 0 148 0 0.8 1 0 normal No
67 1 asymptomatic 125 254 1 0 163 0 0.2 2 2 reversable Yes
62 1 asymptomatic 120 267 0 0 99 1 1.8 2 2 reversable Yes
65 1 asymptomatic 110 248 0 2 158 0 0.6 1 2 fixed Yes
44 1 asymptomatic 110 197 0 2 177 0 0.0 1 1 normal Yes
65 0 nonanginal 160 360 0 2 151 0 0.8 1 0 normal No
60 1 asymptomatic 125 258 0 2 141 1 2.8 2 1 reversable Yes
51 0 nonanginal 140 308 0 2 142 0 1.5 1 1 normal No
48 1 nontypical 130 245 0 2 180 0 0.2 2 0 normal No
58 1 asymptomatic 150 270 0 2 111 1 0.8 1 0 reversable Yes
45 1 asymptomatic 104 208 0 2 148 1 3.0 2 0 normal No
53 0 asymptomatic 130 264 0 2 143 0 0.4 2 0 normal No
39 1 nonanginal 140 321 0 2 182 0 0.0 1 0 normal No
68 1 nonanginal 180 274 1 2 150 1 1.6 2 0 reversable Yes
52 1 nontypical 120 325 0 0 172 0 0.2 1 0 normal No
44 1 nonanginal 140 235 0 2 180 0 0.0 1 0 normal No
47 1 nonanginal 138 257 0 2 156 0 0.0 1 0 normal No
53 0 nonanginal 128 216 0 2 115 0 0.0 1 0 NA No
53 0 asymptomatic 138 234 0 2 160 0 0.0 1 0 normal No
51 0 nonanginal 130 256 0 2 149 0 0.5 1 0 normal No
66 1 asymptomatic 120 302 0 2 151 0 0.4 2 0 normal No
62 0 asymptomatic 160 164 0 2 145 0 6.2 3 3 reversable Yes
62 1 nonanginal 130 231 0 0 146 0 1.8 2 3 reversable No
44 0 nonanginal 108 141 0 0 175 0 0.6 2 0 normal No
63 0 nonanginal 135 252 0 2 172 0 0.0 1 0 normal No
52 1 asymptomatic 128 255 0 0 161 1 0.0 1 1 reversable Yes
59 1 asymptomatic 110 239 0 2 142 1 1.2 2 1 reversable Yes
60 0 asymptomatic 150 258 0 2 157 0 2.6 2 2 reversable Yes
52 1 nontypical 134 201 0 0 158 0 0.8 1 1 normal No
48 1 asymptomatic 122 222 0 2 186 0 0.0 1 0 normal No
45 1 asymptomatic 115 260 0 2 185 0 0.0 1 0 normal No
34 1 typical 118 182 0 2 174 0 0.0 1 0 normal No
57 0 asymptomatic 128 303 0 2 159 0 0.0 1 1 normal No
71 0 nonanginal 110 265 1 2 130 0 0.0 1 1 normal No
49 1 nonanginal 120 188 0 0 139 0 2.0 2 3 reversable Yes
54 1 nontypical 108 309 0 0 156 0 0.0 1 0 reversable No
59 1 asymptomatic 140 177 0 0 162 1 0.0 1 1 reversable Yes
57 1 nonanginal 128 229 0 2 150 0 0.4 2 1 reversable Yes
61 1 asymptomatic 120 260 0 0 140 1 3.6 2 1 reversable Yes
39 1 asymptomatic 118 219 0 0 140 0 1.2 2 0 reversable Yes
61 0 asymptomatic 145 307 0 2 146 1 1.0 2 0 reversable Yes
56 1 asymptomatic 125 249 1 2 144 1 1.2 2 1 normal Yes
52 1 typical 118 186 0 2 190 0 0.0 2 0 fixed No
43 0 asymptomatic 132 341 1 2 136 1 3.0 2 0 reversable Yes
62 0 nonanginal 130 263 0 0 97 0 1.2 2 1 reversable Yes
41 1 nontypical 135 203 0 0 132 0 0.0 2 0 fixed No
58 1 nonanginal 140 211 1 2 165 0 0.0 1 0 normal No
35 0 asymptomatic 138 183 0 0 182 0 1.4 1 0 normal No
63 1 asymptomatic 130 330 1 2 132 1 1.8 1 3 reversable Yes
65 1 asymptomatic 135 254 0 2 127 0 2.8 2 1 reversable Yes
48 1 asymptomatic 130 256 1 2 150 1 0.0 1 2 reversable Yes
63 0 asymptomatic 150 407 0 2 154 0 4.0 2 3 reversable Yes
51 1 nonanginal 100 222 0 0 143 1 1.2 2 0 normal No
55 1 asymptomatic 140 217 0 0 111 1 5.6 3 0 reversable Yes
65 1 typical 138 282 1 2 174 0 1.4 2 1 normal Yes
45 0 nontypical 130 234 0 2 175 0 0.6 2 0 normal No
56 0 asymptomatic 200 288 1 2 133 1 4.0 3 2 reversable Yes
54 1 asymptomatic 110 239 0 0 126 1 2.8 2 1 reversable Yes
44 1 nontypical 120 220 0 0 170 0 0.0 1 0 normal No
62 0 asymptomatic 124 209 0 0 163 0 0.0 1 0 normal No
54 1 nonanginal 120 258 0 2 147 0 0.4 2 0 reversable No
51 1 nonanginal 94 227 0 0 154 1 0.0 1 1 reversable No
29 1 nontypical 130 204 0 2 202 0 0.0 1 0 normal No
51 1 asymptomatic 140 261 0 2 186 1 0.0 1 0 normal No
43 0 nonanginal 122 213 0 0 165 0 0.2 2 0 normal No
55 0 nontypical 135 250 0 2 161 0 1.4 2 0 normal No
70 1 asymptomatic 145 174 0 0 125 1 2.6 3 0 reversable Yes
62 1 nontypical 120 281 0 2 103 0 1.4 2 1 reversable Yes
35 1 asymptomatic 120 198 0 0 130 1 1.6 2 0 reversable Yes
51 1 nonanginal 125 245 1 2 166 0 2.4 2 0 normal No
59 1 nontypical 140 221 0 0 164 1 0.0 1 0 normal No
59 1 typical 170 288 0 2 159 0 0.2 2 0 reversable Yes
52 1 nontypical 128 205 1 0 184 0 0.0 1 0 normal No
64 1 nonanginal 125 309 0 0 131 1 1.8 2 0 reversable Yes
58 1 nonanginal 105 240 0 2 154 1 0.6 2 0 reversable No
47 1 nonanginal 108 243 0 0 152 0 0.0 1 0 normal Yes
57 1 asymptomatic 165 289 1 2 124 0 1.0 2 3 reversable Yes
41 1 nonanginal 112 250 0 0 179 0 0.0 1 0 normal No
45 1 nontypical 128 308 0 2 170 0 0.0 1 0 normal No
60 0 nonanginal 102 318 0 0 160 0 0.0 1 1 normal No
52 1 typical 152 298 1 0 178 0 1.2 2 0 reversable No
42 0 asymptomatic 102 265 0 2 122 0 0.6 2 0 normal No
67 0 nonanginal 115 564 0 2 160 0 1.6 2 0 reversable No
55 1 asymptomatic 160 289 0 2 145 1 0.8 2 1 reversable Yes
64 1 asymptomatic 120 246 0 2 96 1 2.2 3 1 normal Yes
70 1 asymptomatic 130 322 0 2 109 0 2.4 2 3 normal Yes
51 1 asymptomatic 140 299 0 0 173 1 1.6 1 0 reversable Yes
58 1 asymptomatic 125 300 0 2 171 0 0.0 1 2 reversable Yes
60 1 asymptomatic 140 293 0 2 170 0 1.2 2 2 reversable Yes
68 1 nonanginal 118 277 0 0 151 0 1.0 1 1 reversable No
46 1 nontypical 101 197 1 0 156 0 0.0 1 0 reversable No
77 1 asymptomatic 125 304 0 2 162 1 0.0 1 3 normal Yes
54 0 nonanginal 110 214 0 0 158 0 1.6 2 0 normal No
58 0 asymptomatic 100 248 0 2 122 0 1.0 2 0 normal No
48 1 nonanginal 124 255 1 0 175 0 0.0 1 2 normal No
57 1 asymptomatic 132 207 0 0 168 1 0.0 1 0 reversable No
52 1 nonanginal 138 223 0 0 169 0 0.0 1 NA normal No
54 0 nontypical 132 288 1 2 159 1 0.0 1 1 normal No
35 1 asymptomatic 126 282 0 2 156 1 0.0 1 0 reversable Yes
45 0 nontypical 112 160 0 0 138 0 0.0 2 0 normal No
70 1 nonanginal 160 269 0 0 112 1 2.9 2 1 reversable Yes
53 1 asymptomatic 142 226 0 2 111 1 0.0 1 0 reversable No
59 0 asymptomatic 174 249 0 0 143 1 0.0 2 0 normal Yes
62 0 asymptomatic 140 394 0 2 157 0 1.2 2 0 normal No
64 1 asymptomatic 145 212 0 2 132 0 2.0 2 2 fixed Yes
57 1 asymptomatic 152 274 0 0 88 1 1.2 2 1 reversable Yes
52 1 asymptomatic 108 233 1 0 147 0 0.1 1 3 reversable No
56 1 asymptomatic 132 184 0 2 105 1 2.1 2 1 fixed Yes
43 1 nonanginal 130 315 0 0 162 0 1.9 1 1 normal No
53 1 nonanginal 130 246 1 2 173 0 0.0 1 3 normal No
48 1 asymptomatic 124 274 0 2 166 0 0.5 2 0 reversable Yes
56 0 asymptomatic 134 409 0 2 150 1 1.9 2 2 reversable Yes
42 1 typical 148 244 0 2 178 0 0.8 1 2 normal No
59 1 typical 178 270 0 2 145 0 4.2 3 0 reversable No
60 0 asymptomatic 158 305 0 2 161 0 0.0 1 0 normal Yes
63 0 nontypical 140 195 0 0 179 0 0.0 1 2 normal No
42 1 nonanginal 120 240 1 0 194 0 0.8 3 0 reversable No
66 1 nontypical 160 246 0 0 120 1 0.0 2 3 fixed Yes
54 1 nontypical 192 283 0 2 195 0 0.0 1 1 reversable Yes
69 1 nonanginal 140 254 0 2 146 0 2.0 2 3 reversable Yes
50 1 nonanginal 129 196 0 0 163 0 0.0 1 0 normal No
51 1 asymptomatic 140 298 0 0 122 1 4.2 2 3 reversable Yes
43 1 asymptomatic 132 247 1 2 143 1 0.1 2 NA reversable Yes
62 0 asymptomatic 138 294 1 0 106 0 1.9 2 3 normal Yes
68 0 nonanginal 120 211 0 2 115 0 1.5 2 0 normal No
67 1 asymptomatic 100 299 0 2 125 1 0.9 2 2 normal Yes
69 1 typical 160 234 1 2 131 0 0.1 2 1 normal No
45 0 asymptomatic 138 236 0 2 152 1 0.2 2 0 normal No
50 0 nontypical 120 244 0 0 162 0 1.1 1 0 normal No
59 1 typical 160 273 0 2 125 0 0.0 1 0 normal Yes
50 0 asymptomatic 110 254 0 2 159 0 0.0 1 0 normal No
64 0 asymptomatic 180 325 0 0 154 1 0.0 1 0 normal No
57 1 nonanginal 150 126 1 0 173 0 0.2 1 1 reversable No
64 0 nonanginal 140 313 0 0 133 0 0.2 1 0 reversable No
43 1 asymptomatic 110 211 0 0 161 0 0.0 1 0 reversable No
45 1 asymptomatic 142 309 0 2 147 1 0.0 2 3 reversable Yes
58 1 asymptomatic 128 259 0 2 130 1 3.0 2 2 reversable Yes
50 1 asymptomatic 144 200 0 2 126 1 0.9 2 0 reversable Yes
55 1 nontypical 130 262 0 0 155 0 0.0 1 0 normal No
62 0 asymptomatic 150 244 0 0 154 1 1.4 2 0 normal Yes
37 0 nonanginal 120 215 0 0 170 0 0.0 1 0 normal No
38 1 typical 120 231 0 0 182 1 3.8 2 0 reversable Yes
41 1 nonanginal 130 214 0 2 168 0 2.0 2 0 normal No
66 0 asymptomatic 178 228 1 0 165 1 1.0 2 2 reversable Yes
52 1 asymptomatic 112 230 0 0 160 0 0.0 1 1 normal Yes
56 1 typical 120 193 0 2 162 0 1.9 2 0 reversable No
46 0 nontypical 105 204 0 0 172 0 0.0 1 0 normal No
46 0 asymptomatic 138 243 0 2 152 1 0.0 2 0 normal No
64 0 asymptomatic 130 303 0 0 122 0 2.0 2 2 normal No
59 1 asymptomatic 138 271 0 2 182 0 0.0 1 0 normal No
41 0 nonanginal 112 268 0 2 172 1 0.0 1 0 normal No
54 0 nonanginal 108 267 0 2 167 0 0.0 1 0 normal No
39 0 nonanginal 94 199 0 0 179 0 0.0 1 0 normal No
53 1 asymptomatic 123 282 0 0 95 1 2.0 2 2 reversable Yes
63 0 asymptomatic 108 269 0 0 169 1 1.8 2 2 normal Yes
34 0 nontypical 118 210 0 0 192 0 0.7 1 0 normal No
47 1 asymptomatic 112 204 0 0 143 0 0.1 1 0 normal No
67 0 nonanginal 152 277 0 0 172 0 0.0 1 1 normal No
54 1 asymptomatic 110 206 0 2 108 1 0.0 2 1 normal Yes
66 1 asymptomatic 112 212 0 2 132 1 0.1 1 1 normal Yes
52 0 nonanginal 136 196 0 2 169 0 0.1 2 0 normal No
55 0 asymptomatic 180 327 0 1 117 1 3.4 2 0 normal Yes
49 1 nonanginal 118 149 0 2 126 0 0.8 1 3 normal Yes
74 0 nontypical 120 269 0 2 121 1 0.2 1 1 normal No
54 0 nonanginal 160 201 0 0 163 0 0.0 1 1 normal No
54 1 asymptomatic 122 286 0 2 116 1 3.2 2 2 normal Yes
56 1 asymptomatic 130 283 1 2 103 1 1.6 3 0 reversable Yes
46 1 asymptomatic 120 249 0 2 144 0 0.8 1 0 reversable Yes
49 0 nontypical 134 271 0 0 162 0 0.0 2 0 normal No
42 1 nontypical 120 295 0 0 162 0 0.0 1 0 normal No
41 1 nontypical 110 235 0 0 153 0 0.0 1 0 normal No
41 0 nontypical 126 306 0 0 163 0 0.0 1 0 normal No
49 0 asymptomatic 130 269 0 0 163 0 0.0 1 0 normal No
61 1 typical 134 234 0 0 145 0 2.6 2 2 normal Yes
60 0 nonanginal 120 178 1 0 96 0 0.0 1 0 normal No
67 1 asymptomatic 120 237 0 0 71 0 1.0 2 0 normal Yes
58 1 asymptomatic 100 234 0 0 156 0 0.1 1 1 reversable Yes
47 1 asymptomatic 110 275 0 2 118 1 1.0 2 1 normal Yes
52 1 asymptomatic 125 212 0 0 168 0 1.0 1 2 reversable Yes
62 1 nontypical 128 208 1 2 140 0 0.0 1 0 normal No
57 1 asymptomatic 110 201 0 0 126 1 1.5 2 0 fixed No
58 1 asymptomatic 146 218 0 0 105 0 2.0 2 1 reversable Yes
64 1 asymptomatic 128 263 0 0 105 1 0.2 2 1 reversable No
51 0 nonanginal 120 295 0 2 157 0 0.6 1 0 normal No
43 1 asymptomatic 115 303 0 0 181 0 1.2 2 0 normal No
42 0 nonanginal 120 209 0 0 173 0 0.0 2 0 normal No
67 0 asymptomatic 106 223 0 0 142 0 0.3 1 2 normal No
76 0 nonanginal 140 197 0 1 116 0 1.1 2 0 normal No
70 1 nontypical 156 245 0 2 143 0 0.0 1 0 normal No
57 1 nontypical 124 261 0 0 141 0 0.3 1 0 reversable Yes
44 0 nonanginal 118 242 0 0 149 0 0.3 2 1 normal No
58 0 nontypical 136 319 1 2 152 0 0.0 1 2 normal Yes
60 0 typical 150 240 0 0 171 0 0.9 1 0 normal No
44 1 nonanginal 120 226 0 0 169 0 0.0 1 0 normal No
61 1 asymptomatic 138 166 0 2 125 1 3.6 2 1 normal Yes
42 1 asymptomatic 136 315 0 0 125 1 1.8 2 0 fixed Yes
52 1 asymptomatic 128 204 1 0 156 1 1.0 2 0 NA Yes
59 1 nonanginal 126 218 1 0 134 0 2.2 2 1 fixed Yes
40 1 asymptomatic 152 223 0 0 181 0 0.0 1 0 reversable Yes
42 1 nonanginal 130 180 0 0 150 0 0.0 1 0 normal No
61 1 asymptomatic 140 207 0 2 138 1 1.9 1 1 reversable Yes
66 1 asymptomatic 160 228 0 2 138 0 2.3 1 0 fixed No
46 1 asymptomatic 140 311 0 0 120 1 1.8 2 2 reversable Yes
71 0 asymptomatic 112 149 0 0 125 0 1.6 2 0 normal No
59 1 typical 134 204 0 0 162 0 0.8 1 2 normal Yes
64 1 typical 170 227 0 2 155 0 0.6 2 0 reversable No
66 0 nonanginal 146 278 0 2 152 0 0.0 2 1 normal No
39 0 nonanginal 138 220 0 0 152 0 0.0 2 0 normal No
57 1 nontypical 154 232 0 2 164 0 0.0 1 1 normal Yes
58 0 asymptomatic 130 197 0 0 131 0 0.6 2 0 normal No
57 1 asymptomatic 110 335 0 0 143 1 3.0 2 1 reversable Yes
47 1 nonanginal 130 253 0 0 179 0 0.0 1 0 normal No
55 0 asymptomatic 128 205 0 1 130 1 2.0 2 1 reversable Yes
35 1 nontypical 122 192 0 0 174 0 0.0 1 0 normal No
61 1 asymptomatic 148 203 0 0 161 0 0.0 1 1 reversable Yes
58 1 asymptomatic 114 318 0 1 140 0 4.4 3 3 fixed Yes
58 0 asymptomatic 170 225 1 2 146 1 2.8 2 2 fixed Yes
58 1 nontypical 125 220 0 0 144 0 0.4 2 NA reversable No
56 1 nontypical 130 221 0 2 163 0 0.0 1 0 reversable No
56 1 nontypical 120 240 0 0 169 0 0.0 3 0 normal No
67 1 nonanginal 152 212 0 2 150 0 0.8 2 0 reversable Yes
55 0 nontypical 132 342 0 0 166 0 1.2 1 0 normal No
44 1 asymptomatic 120 169 0 0 144 1 2.8 3 0 fixed Yes
63 1 asymptomatic 140 187 0 2 144 1 4.0 1 2 reversable Yes
63 0 asymptomatic 124 197 0 0 136 1 0.0 2 0 normal Yes
41 1 nontypical 120 157 0 0 182 0 0.0 1 0 normal No
59 1 asymptomatic 164 176 1 2 90 0 1.0 2 2 fixed Yes
57 0 asymptomatic 140 241 0 0 123 1 0.2 2 0 reversable Yes
45 1 typical 110 264 0 0 132 0 1.2 2 0 reversable Yes
68 1 asymptomatic 144 193 1 0 141 0 3.4 2 2 reversable Yes
57 1 asymptomatic 130 131 0 0 115 1 1.2 2 1 reversable Yes
57 0 nontypical 130 236 0 2 174 0 0.0 2 1 normal Yes
38 1 nonanginal 138 175 0 0 173 0 0.0 1 NA normal No

38.7.3 CarSeats数据

knitr::kable(Carseats)
Sales CompPrice Income Advertising Population Price ShelveLoc Age Education Urban US
9.50 138 73 11 276 120 Bad 42 17 Yes Yes
11.22 111 48 16 260 83 Good 65 10 Yes Yes
10.06 113 35 10 269 80 Medium 59 12 Yes Yes
7.40 117 100 4 466 97 Medium 55 14 Yes Yes
4.15 141 64 3 340 128 Bad 38 13 Yes No
10.81 124 113 13 501 72 Bad 78 16 No Yes
6.63 115 105 0 45 108 Medium 71 15 Yes No
11.85 136 81 15 425 120 Good 67 10 Yes Yes
6.54 132 110 0 108 124 Medium 76 10 No No
4.69 132 113 0 131 124 Medium 76 17 No Yes
9.01 121 78 9 150 100 Bad 26 10 No Yes
11.96 117 94 4 503 94 Good 50 13 Yes Yes
3.98 122 35 2 393 136 Medium 62 18 Yes No
10.96 115 28 11 29 86 Good 53 18 Yes Yes
11.17 107 117 11 148 118 Good 52 18 Yes Yes
8.71 149 95 5 400 144 Medium 76 18 No No
7.58 118 32 0 284 110 Good 63 13 Yes No
12.29 147 74 13 251 131 Good 52 10 Yes Yes
13.91 110 110 0 408 68 Good 46 17 No Yes
8.73 129 76 16 58 121 Medium 69 12 Yes Yes
6.41 125 90 2 367 131 Medium 35 18 Yes Yes
12.13 134 29 12 239 109 Good 62 18 No Yes
5.08 128 46 6 497 138 Medium 42 13 Yes No
5.87 121 31 0 292 109 Medium 79 10 Yes No
10.14 145 119 16 294 113 Bad 42 12 Yes Yes
14.90 139 32 0 176 82 Good 54 11 No No
8.33 107 115 11 496 131 Good 50 11 No Yes
5.27 98 118 0 19 107 Medium 64 17 Yes No
2.99 103 74 0 359 97 Bad 55 11 Yes Yes
7.81 104 99 15 226 102 Bad 58 17 Yes Yes
13.55 125 94 0 447 89 Good 30 12 Yes No
8.25 136 58 16 241 131 Medium 44 18 Yes Yes
6.20 107 32 12 236 137 Good 64 10 No Yes
8.77 114 38 13 317 128 Good 50 16 Yes Yes
2.67 115 54 0 406 128 Medium 42 17 Yes Yes
11.07 131 84 11 29 96 Medium 44 17 No Yes
8.89 122 76 0 270 100 Good 60 18 No No
4.95 121 41 5 412 110 Medium 54 10 Yes Yes
6.59 109 73 0 454 102 Medium 65 15 Yes No
3.24 130 60 0 144 138 Bad 38 10 No No
2.07 119 98 0 18 126 Bad 73 17 No No
7.96 157 53 0 403 124 Bad 58 16 Yes No
10.43 77 69 0 25 24 Medium 50 18 Yes No
4.12 123 42 11 16 134 Medium 59 13 Yes Yes
4.16 85 79 6 325 95 Medium 69 13 Yes Yes
4.56 141 63 0 168 135 Bad 44 12 Yes Yes
12.44 127 90 14 16 70 Medium 48 15 No Yes
4.38 126 98 0 173 108 Bad 55 16 Yes No
3.91 116 52 0 349 98 Bad 69 18 Yes No
10.61 157 93 0 51 149 Good 32 17 Yes No
1.42 99 32 18 341 108 Bad 80 16 Yes Yes
4.42 121 90 0 150 108 Bad 75 16 Yes No
7.91 153 40 3 112 129 Bad 39 18 Yes Yes
6.92 109 64 13 39 119 Medium 61 17 Yes Yes
4.90 134 103 13 25 144 Medium 76 17 No Yes
6.85 143 81 5 60 154 Medium 61 18 Yes Yes
11.91 133 82 0 54 84 Medium 50 17 Yes No
0.91 93 91 0 22 117 Bad 75 11 Yes No
5.42 103 93 15 188 103 Bad 74 16 Yes Yes
5.21 118 71 4 148 114 Medium 80 13 Yes No
8.32 122 102 19 469 123 Bad 29 13 Yes Yes
7.32 105 32 0 358 107 Medium 26 13 No No
1.82 139 45 0 146 133 Bad 77 17 Yes Yes
8.47 119 88 10 170 101 Medium 61 13 Yes Yes
7.80 100 67 12 184 104 Medium 32 16 No Yes
4.90 122 26 0 197 128 Medium 55 13 No No
8.85 127 92 0 508 91 Medium 56 18 Yes No
9.01 126 61 14 152 115 Medium 47 16 Yes Yes
13.39 149 69 20 366 134 Good 60 13 Yes Yes
7.99 127 59 0 339 99 Medium 65 12 Yes No
9.46 89 81 15 237 99 Good 74 12 Yes Yes
6.50 148 51 16 148 150 Medium 58 17 No Yes
5.52 115 45 0 432 116 Medium 25 15 Yes No
12.61 118 90 10 54 104 Good 31 11 No Yes
6.20 150 68 5 125 136 Medium 64 13 No Yes
8.55 88 111 23 480 92 Bad 36 16 No Yes
10.64 102 87 10 346 70 Medium 64 15 Yes Yes
7.70 118 71 12 44 89 Medium 67 18 No Yes
4.43 134 48 1 139 145 Medium 65 12 Yes Yes
9.14 134 67 0 286 90 Bad 41 13 Yes No
8.01 113 100 16 353 79 Bad 68 11 Yes Yes
7.52 116 72 0 237 128 Good 70 13 Yes No
11.62 151 83 4 325 139 Good 28 17 Yes Yes
4.42 109 36 7 468 94 Bad 56 11 Yes Yes
2.23 111 25 0 52 121 Bad 43 18 No No
8.47 125 103 0 304 112 Medium 49 13 No No
8.70 150 84 9 432 134 Medium 64 15 Yes No
11.70 131 67 7 272 126 Good 54 16 No Yes
6.56 117 42 7 144 111 Medium 62 10 Yes Yes
7.95 128 66 3 493 119 Medium 45 16 No No
5.33 115 22 0 491 103 Medium 64 11 No No
4.81 97 46 11 267 107 Medium 80 15 Yes Yes
4.53 114 113 0 97 125 Medium 29 12 Yes No
8.86 145 30 0 67 104 Medium 55 17 Yes No
8.39 115 97 5 134 84 Bad 55 11 Yes Yes
5.58 134 25 10 237 148 Medium 59 13 Yes Yes
9.48 147 42 10 407 132 Good 73 16 No Yes
7.45 161 82 5 287 129 Bad 33 16 Yes Yes
12.49 122 77 24 382 127 Good 36 16 No Yes
4.88 121 47 3 220 107 Bad 56 16 No Yes
4.11 113 69 11 94 106 Medium 76 12 No Yes
6.20 128 93 0 89 118 Medium 34 18 Yes No
5.30 113 22 0 57 97 Medium 65 16 No No
5.07 123 91 0 334 96 Bad 78 17 Yes Yes
4.62 121 96 0 472 138 Medium 51 12 Yes No
5.55 104 100 8 398 97 Medium 61 11 Yes Yes
0.16 102 33 0 217 139 Medium 70 18 No No
8.55 134 107 0 104 108 Medium 60 12 Yes No
3.47 107 79 2 488 103 Bad 65 16 Yes No
8.98 115 65 0 217 90 Medium 60 17 No No
9.00 128 62 7 125 116 Medium 43 14 Yes Yes
6.62 132 118 12 272 151 Medium 43 14 Yes Yes
6.67 116 99 5 298 125 Good 62 12 Yes Yes
6.01 131 29 11 335 127 Bad 33 12 Yes Yes
9.31 122 87 9 17 106 Medium 65 13 Yes Yes
8.54 139 35 0 95 129 Medium 42 13 Yes No
5.08 135 75 0 202 128 Medium 80 10 No No
8.80 145 53 0 507 119 Medium 41 12 Yes No
7.57 112 88 2 243 99 Medium 62 11 Yes Yes
7.37 130 94 8 137 128 Medium 64 12 Yes Yes
6.87 128 105 11 249 131 Medium 63 13 Yes Yes
11.67 125 89 10 380 87 Bad 28 10 Yes Yes
6.88 119 100 5 45 108 Medium 75 10 Yes Yes
8.19 127 103 0 125 155 Good 29 15 No Yes
8.87 131 113 0 181 120 Good 63 14 Yes No
9.34 89 78 0 181 49 Medium 43 15 No No
11.27 153 68 2 60 133 Good 59 16 Yes Yes
6.52 125 48 3 192 116 Medium 51 14 Yes Yes
4.96 133 100 3 350 126 Bad 55 13 Yes Yes
4.47 143 120 7 279 147 Bad 40 10 No Yes
8.41 94 84 13 497 77 Medium 51 12 Yes Yes
6.50 108 69 3 208 94 Medium 77 16 Yes No
9.54 125 87 9 232 136 Good 72 10 Yes Yes
7.62 132 98 2 265 97 Bad 62 12 Yes Yes
3.67 132 31 0 327 131 Medium 76 16 Yes No
6.44 96 94 14 384 120 Medium 36 18 No Yes
5.17 131 75 0 10 120 Bad 31 18 No No
6.52 128 42 0 436 118 Medium 80 11 Yes No
10.27 125 103 12 371 109 Medium 44 10 Yes Yes
12.30 146 62 10 310 94 Medium 30 13 No Yes
6.03 133 60 10 277 129 Medium 45 18 Yes Yes
6.53 140 42 0 331 131 Bad 28 15 Yes No
7.44 124 84 0 300 104 Medium 77 15 Yes No
0.53 122 88 7 36 159 Bad 28 17 Yes Yes
9.09 132 68 0 264 123 Good 34 11 No No
8.77 144 63 11 27 117 Medium 47 17 Yes Yes
3.90 114 83 0 412 131 Bad 39 14 Yes No
10.51 140 54 9 402 119 Good 41 16 No Yes
7.56 110 119 0 384 97 Medium 72 14 No Yes
11.48 121 120 13 140 87 Medium 56 11 Yes Yes
10.49 122 84 8 176 114 Good 57 10 No Yes
10.77 111 58 17 407 103 Good 75 17 No Yes
7.64 128 78 0 341 128 Good 45 13 No No
5.93 150 36 7 488 150 Medium 25 17 No Yes
6.89 129 69 10 289 110 Medium 50 16 No Yes
7.71 98 72 0 59 69 Medium 65 16 Yes No
7.49 146 34 0 220 157 Good 51 16 Yes No
10.21 121 58 8 249 90 Medium 48 13 No Yes
12.53 142 90 1 189 112 Good 39 10 No Yes
9.32 119 60 0 372 70 Bad 30 18 No No
4.67 111 28 0 486 111 Medium 29 12 No No
2.93 143 21 5 81 160 Medium 67 12 No Yes
3.63 122 74 0 424 149 Medium 51 13 Yes No
5.68 130 64 0 40 106 Bad 39 17 No No
8.22 148 64 0 58 141 Medium 27 13 No Yes
0.37 147 58 7 100 191 Bad 27 15 Yes Yes
6.71 119 67 17 151 137 Medium 55 11 Yes Yes
6.71 106 73 0 216 93 Medium 60 13 Yes No
7.30 129 89 0 425 117 Medium 45 10 Yes No
11.48 104 41 15 492 77 Good 73 18 Yes Yes
8.01 128 39 12 356 118 Medium 71 10 Yes Yes
12.49 93 106 12 416 55 Medium 75 15 Yes Yes
9.03 104 102 13 123 110 Good 35 16 Yes Yes
6.38 135 91 5 207 128 Medium 66 18 Yes Yes
0.00 139 24 0 358 185 Medium 79 15 No No
7.54 115 89 0 38 122 Medium 25 12 Yes No
5.61 138 107 9 480 154 Medium 47 11 No Yes
10.48 138 72 0 148 94 Medium 27 17 Yes Yes
10.66 104 71 14 89 81 Medium 25 14 No Yes
7.78 144 25 3 70 116 Medium 77 18 Yes Yes
4.94 137 112 15 434 149 Bad 66 13 Yes Yes
7.43 121 83 0 79 91 Medium 68 11 Yes No
4.74 137 60 4 230 140 Bad 25 13 Yes No
5.32 118 74 6 426 102 Medium 80 18 Yes Yes
9.95 132 33 7 35 97 Medium 60 11 No Yes
10.07 130 100 11 449 107 Medium 64 10 Yes Yes
8.68 120 51 0 93 86 Medium 46 17 No No
6.03 117 32 0 142 96 Bad 62 17 Yes No
8.07 116 37 0 426 90 Medium 76 15 Yes No
12.11 118 117 18 509 104 Medium 26 15 No Yes
8.79 130 37 13 297 101 Medium 37 13 No Yes
6.67 156 42 13 170 173 Good 74 14 Yes Yes
7.56 108 26 0 408 93 Medium 56 14 No No
13.28 139 70 7 71 96 Good 61 10 Yes Yes
7.23 112 98 18 481 128 Medium 45 11 Yes Yes
4.19 117 93 4 420 112 Bad 66 11 Yes Yes
4.10 130 28 6 410 133 Bad 72 16 Yes Yes
2.52 124 61 0 333 138 Medium 76 16 Yes No
3.62 112 80 5 500 128 Medium 69 10 Yes Yes
6.42 122 88 5 335 126 Medium 64 14 Yes Yes
5.56 144 92 0 349 146 Medium 62 12 No No
5.94 138 83 0 139 134 Medium 54 18 Yes No
4.10 121 78 4 413 130 Bad 46 10 No Yes
2.05 131 82 0 132 157 Bad 25 14 Yes No
8.74 155 80 0 237 124 Medium 37 14 Yes No
5.68 113 22 1 317 132 Medium 28 12 Yes No
4.97 162 67 0 27 160 Medium 77 17 Yes Yes
8.19 111 105 0 466 97 Bad 61 10 No No
7.78 86 54 0 497 64 Bad 33 12 Yes No
3.02 98 21 11 326 90 Bad 76 11 No Yes
4.36 125 41 2 357 123 Bad 47 14 No Yes
9.39 117 118 14 445 120 Medium 32 15 Yes Yes
12.04 145 69 19 501 105 Medium 45 11 Yes Yes
8.23 149 84 5 220 139 Medium 33 10 Yes Yes
4.83 115 115 3 48 107 Medium 73 18 Yes Yes
2.34 116 83 15 170 144 Bad 71 11 Yes Yes
5.73 141 33 0 243 144 Medium 34 17 Yes No
4.34 106 44 0 481 111 Medium 70 14 No No
9.70 138 61 12 156 120 Medium 25 14 Yes Yes
10.62 116 79 19 359 116 Good 58 17 Yes Yes
10.59 131 120 15 262 124 Medium 30 10 Yes Yes
6.43 124 44 0 125 107 Medium 80 11 Yes No
7.49 136 119 6 178 145 Medium 35 13 Yes Yes
3.45 110 45 9 276 125 Medium 62 14 Yes Yes
4.10 134 82 0 464 141 Medium 48 13 No No
6.68 107 25 0 412 82 Bad 36 14 Yes No
7.80 119 33 0 245 122 Good 56 14 Yes No
8.69 113 64 10 68 101 Medium 57 16 Yes Yes
5.40 149 73 13 381 163 Bad 26 11 No Yes
11.19 98 104 0 404 72 Medium 27 18 No No
5.16 115 60 0 119 114 Bad 38 14 No No
8.09 132 69 0 123 122 Medium 27 11 No No
13.14 137 80 10 24 105 Good 61 15 Yes Yes
8.65 123 76 18 218 120 Medium 29 14 No Yes
9.43 115 62 11 289 129 Good 56 16 No Yes
5.53 126 32 8 95 132 Medium 50 17 Yes Yes
9.32 141 34 16 361 108 Medium 69 10 Yes Yes
9.62 151 28 8 499 135 Medium 48 10 Yes Yes
7.36 121 24 0 200 133 Good 73 13 Yes No
3.89 123 105 0 149 118 Bad 62 16 Yes Yes
10.31 159 80 0 362 121 Medium 26 18 Yes No
12.01 136 63 0 160 94 Medium 38 12 Yes No
4.68 124 46 0 199 135 Medium 52 14 No No
7.82 124 25 13 87 110 Medium 57 10 Yes Yes
8.78 130 30 0 391 100 Medium 26 18 Yes No
10.00 114 43 0 199 88 Good 57 10 No Yes
6.90 120 56 20 266 90 Bad 78 18 Yes Yes
5.04 123 114 0 298 151 Bad 34 16 Yes No
5.36 111 52 0 12 101 Medium 61 11 Yes Yes
5.05 125 67 0 86 117 Bad 65 11 Yes No
9.16 137 105 10 435 156 Good 72 14 Yes Yes
3.72 139 111 5 310 132 Bad 62 13 Yes Yes
8.31 133 97 0 70 117 Medium 32 16 Yes No
5.64 124 24 5 288 122 Medium 57 12 No Yes
9.58 108 104 23 353 129 Good 37 17 Yes Yes
7.71 123 81 8 198 81 Bad 80 15 Yes Yes
4.20 147 40 0 277 144 Medium 73 10 Yes No
8.67 125 62 14 477 112 Medium 80 13 Yes Yes
3.47 108 38 0 251 81 Bad 72 14 No No
5.12 123 36 10 467 100 Bad 74 11 No Yes
7.67 129 117 8 400 101 Bad 36 10 Yes Yes
5.71 121 42 4 188 118 Medium 54 15 Yes Yes
6.37 120 77 15 86 132 Medium 48 18 Yes Yes
7.77 116 26 6 434 115 Medium 25 17 Yes Yes
6.95 128 29 5 324 159 Good 31 15 Yes Yes
5.31 130 35 10 402 129 Bad 39 17 Yes Yes
9.10 128 93 12 343 112 Good 73 17 No Yes
5.83 134 82 7 473 112 Bad 51 12 No Yes
6.53 123 57 0 66 105 Medium 39 11 Yes No
5.01 159 69 0 438 166 Medium 46 17 Yes No
11.99 119 26 0 284 89 Good 26 10 Yes No
4.55 111 56 0 504 110 Medium 62 16 Yes No
12.98 113 33 0 14 63 Good 38 12 Yes No
10.04 116 106 8 244 86 Medium 58 12 Yes Yes
7.22 135 93 2 67 119 Medium 34 11 Yes Yes
6.67 107 119 11 210 132 Medium 53 11 Yes Yes
6.93 135 69 14 296 130 Medium 73 15 Yes Yes
7.80 136 48 12 326 125 Medium 36 16 Yes Yes
7.22 114 113 2 129 151 Good 40 15 No Yes
3.42 141 57 13 376 158 Medium 64 18 Yes Yes
2.86 121 86 10 496 145 Bad 51 10 Yes Yes
11.19 122 69 7 303 105 Good 45 16 No Yes
7.74 150 96 0 80 154 Good 61 11 Yes No
5.36 135 110 0 112 117 Medium 80 16 No No
6.97 106 46 11 414 96 Bad 79 17 No No
7.60 146 26 11 261 131 Medium 39 10 Yes Yes
7.53 117 118 11 429 113 Medium 67 18 No Yes
6.88 95 44 4 208 72 Bad 44 17 Yes Yes
6.98 116 40 0 74 97 Medium 76 15 No No
8.75 143 77 25 448 156 Medium 43 17 Yes Yes
9.49 107 111 14 400 103 Medium 41 11 No Yes
6.64 118 70 0 106 89 Bad 39 17 Yes No
11.82 113 66 16 322 74 Good 76 15 Yes Yes
11.28 123 84 0 74 89 Good 59 10 Yes No
12.66 148 76 3 126 99 Good 60 11 Yes Yes
4.21 118 35 14 502 137 Medium 79 10 No Yes
8.21 127 44 13 160 123 Good 63 18 Yes Yes
3.07 118 83 13 276 104 Bad 75 10 Yes Yes
10.98 148 63 0 312 130 Good 63 15 Yes No
9.40 135 40 17 497 96 Medium 54 17 No Yes
8.57 116 78 1 158 99 Medium 45 11 Yes Yes
7.41 99 93 0 198 87 Medium 57 16 Yes Yes
5.28 108 77 13 388 110 Bad 74 14 Yes Yes
10.01 133 52 16 290 99 Medium 43 11 Yes Yes
11.93 123 98 12 408 134 Good 29 10 Yes Yes
8.03 115 29 26 394 132 Medium 33 13 Yes Yes
4.78 131 32 1 85 133 Medium 48 12 Yes Yes
5.90 138 92 0 13 120 Bad 61 12 Yes No
9.24 126 80 19 436 126 Medium 52 10 Yes Yes
11.18 131 111 13 33 80 Bad 68 18 Yes Yes
9.53 175 65 29 419 166 Medium 53 12 Yes Yes
6.15 146 68 12 328 132 Bad 51 14 Yes Yes
6.80 137 117 5 337 135 Bad 38 10 Yes Yes
9.33 103 81 3 491 54 Medium 66 13 Yes No
7.72 133 33 10 333 129 Good 71 14 Yes Yes
6.39 131 21 8 220 171 Good 29 14 Yes Yes
15.63 122 36 5 369 72 Good 35 10 Yes Yes
6.41 142 30 0 472 136 Good 80 15 No No
10.08 116 72 10 456 130 Good 41 14 No Yes
6.97 127 45 19 459 129 Medium 57 11 No Yes
5.86 136 70 12 171 152 Medium 44 18 Yes Yes
7.52 123 39 5 499 98 Medium 34 15 Yes No
9.16 140 50 10 300 139 Good 60 15 Yes Yes
10.36 107 105 18 428 103 Medium 34 12 Yes Yes
2.66 136 65 4 133 150 Bad 53 13 Yes Yes
11.70 144 69 11 131 104 Medium 47 11 Yes Yes
4.69 133 30 0 152 122 Medium 53 17 Yes No
6.23 112 38 17 316 104 Medium 80 16 Yes Yes
3.15 117 66 1 65 111 Bad 55 11 Yes Yes
11.27 100 54 9 433 89 Good 45 12 Yes Yes
4.99 122 59 0 501 112 Bad 32 14 No No
10.10 135 63 15 213 134 Medium 32 10 Yes Yes
5.74 106 33 20 354 104 Medium 61 12 Yes Yes
5.87 136 60 7 303 147 Medium 41 10 Yes Yes
7.63 93 117 9 489 83 Bad 42 13 Yes Yes
6.18 120 70 15 464 110 Medium 72 15 Yes Yes
5.17 138 35 6 60 143 Bad 28 18 Yes No
8.61 130 38 0 283 102 Medium 80 15 Yes No
5.97 112 24 0 164 101 Medium 45 11 Yes No
11.54 134 44 4 219 126 Good 44 15 Yes Yes
7.50 140 29 0 105 91 Bad 43 16 Yes No
7.38 98 120 0 268 93 Medium 72 10 No No
7.81 137 102 13 422 118 Medium 71 10 No Yes
5.99 117 42 10 371 121 Bad 26 14 Yes Yes
8.43 138 80 0 108 126 Good 70 13 No Yes
4.81 121 68 0 279 149 Good 79 12 Yes No
8.97 132 107 0 144 125 Medium 33 13 No No
6.88 96 39 0 161 112 Good 27 14 No No
12.57 132 102 20 459 107 Good 49 11 Yes Yes
9.32 134 27 18 467 96 Medium 49 14 No Yes
8.64 111 101 17 266 91 Medium 63 17 No Yes
10.44 124 115 16 458 105 Medium 62 16 No Yes
13.44 133 103 14 288 122 Good 61 17 Yes Yes
9.45 107 67 12 430 92 Medium 35 12 No Yes
5.30 133 31 1 80 145 Medium 42 18 Yes Yes
7.02 130 100 0 306 146 Good 42 11 Yes No
3.58 142 109 0 111 164 Good 72 12 Yes No
13.36 103 73 3 276 72 Medium 34 15 Yes Yes
4.17 123 96 10 71 118 Bad 69 11 Yes Yes
3.13 130 62 11 396 130 Bad 66 14 Yes Yes
8.77 118 86 7 265 114 Good 52 15 No Yes
8.68 131 25 10 183 104 Medium 56 15 No Yes
5.25 131 55 0 26 110 Bad 79 12 Yes Yes
10.26 111 75 1 377 108 Good 25 12 Yes No
10.50 122 21 16 488 131 Good 30 14 Yes Yes
6.53 154 30 0 122 162 Medium 57 17 No No
5.98 124 56 11 447 134 Medium 53 12 No Yes
14.37 95 106 0 256 53 Good 52 17 Yes No
10.71 109 22 10 348 79 Good 74 14 No Yes
10.26 135 100 22 463 122 Medium 36 14 Yes Yes
7.68 126 41 22 403 119 Bad 42 12 Yes Yes
9.08 152 81 0 191 126 Medium 54 16 Yes No
7.80 121 50 0 508 98 Medium 65 11 No No
5.58 137 71 0 402 116 Medium 78 17 Yes No
9.44 131 47 7 90 118 Medium 47 12 Yes Yes
7.90 132 46 4 206 124 Medium 73 11 Yes No
16.27 141 60 19 319 92 Good 44 11 Yes Yes
6.81 132 61 0 263 125 Medium 41 12 No No
6.11 133 88 3 105 119 Medium 79 12 Yes Yes
5.81 125 111 0 404 107 Bad 54 15 Yes No
9.64 106 64 10 17 89 Medium 68 17 Yes Yes
3.90 124 65 21 496 151 Bad 77 13 Yes Yes
4.95 121 28 19 315 121 Medium 66 14 Yes Yes
9.35 98 117 0 76 68 Medium 63 10 Yes No
12.85 123 37 15 348 112 Good 28 12 Yes Yes
5.87 131 73 13 455 132 Medium 62 17 Yes Yes
5.32 152 116 0 170 160 Medium 39 16 Yes No
8.67 142 73 14 238 115 Medium 73 14 No Yes
8.14 135 89 11 245 78 Bad 79 16 Yes Yes
8.44 128 42 8 328 107 Medium 35 12 Yes Yes
5.47 108 75 9 61 111 Medium 67 12 Yes Yes
6.10 153 63 0 49 124 Bad 56 16 Yes No
4.53 129 42 13 315 130 Bad 34 13 Yes Yes
5.57 109 51 10 26 120 Medium 30 17 No Yes
5.35 130 58 19 366 139 Bad 33 16 Yes Yes
12.57 138 108 17 203 128 Good 33 14 Yes Yes
6.14 139 23 3 37 120 Medium 55 11 No Yes
7.41 162 26 12 368 159 Medium 40 18 Yes Yes
5.94 100 79 7 284 95 Bad 50 12 Yes Yes
9.71 134 37 0 27 120 Good 49 16 Yes Yes

38.7.4 Boston数据

knitr::kable(Boston)
crim zn indus chas nox rm age dis rad tax ptratio black lstat medv
0.00632 18.0 2.31 0 0.5380 6.575 65.2 4.0900 1 296 15.3 396.90 4.98 24.0
0.02731 0.0 7.07 0 0.4690 6.421 78.9 4.9671 2 242 17.8 396.90 9.14 21.6
0.02729 0.0 7.07 0 0.4690 7.185 61.1 4.9671 2 242 17.8 392.83 4.03 34.7
0.03237 0.0 2.18 0 0.4580 6.998 45.8 6.0622 3 222 18.7 394.63 2.94 33.4
0.06905 0.0 2.18 0 0.4580 7.147 54.2 6.0622 3 222 18.7 396.90 5.33 36.2
0.02985 0.0 2.18 0 0.4580 6.430 58.7 6.0622 3 222 18.7 394.12 5.21 28.7
0.08829 12.5 7.87 0 0.5240 6.012 66.6 5.5605 5 311 15.2 395.60 12.43 22.9
0.14455 12.5 7.87 0 0.5240 6.172 96.1 5.9505 5 311 15.2 396.90 19.15 27.1
0.21124 12.5 7.87 0 0.5240 5.631 100.0 6.0821 5 311 15.2 386.63 29.93 16.5
0.17004 12.5 7.87 0 0.5240 6.004 85.9 6.5921 5 311 15.2 386.71 17.10 18.9
0.22489 12.5 7.87 0 0.5240 6.377 94.3 6.3467 5 311 15.2 392.52 20.45 15.0
0.11747 12.5 7.87 0 0.5240 6.009 82.9 6.2267 5 311 15.2 396.90 13.27 18.9
0.09378 12.5 7.87 0 0.5240 5.889 39.0 5.4509 5 311 15.2 390.50 15.71 21.7
0.62976 0.0 8.14 0 0.5380 5.949 61.8 4.7075 4 307 21.0 396.90 8.26 20.4
0.63796 0.0 8.14 0 0.5380 6.096 84.5 4.4619 4 307 21.0 380.02 10.26 18.2
0.62739 0.0 8.14 0 0.5380 5.834 56.5 4.4986 4 307 21.0 395.62 8.47 19.9
1.05393 0.0 8.14 0 0.5380 5.935 29.3 4.4986 4 307 21.0 386.85 6.58 23.1
0.78420 0.0 8.14 0 0.5380 5.990 81.7 4.2579 4 307 21.0 386.75 14.67 17.5
0.80271 0.0 8.14 0 0.5380 5.456 36.6 3.7965 4 307 21.0 288.99 11.69 20.2
0.72580 0.0 8.14 0 0.5380 5.727 69.5 3.7965 4 307 21.0 390.95 11.28 18.2
1.25179 0.0 8.14 0 0.5380 5.570 98.1 3.7979 4 307 21.0 376.57 21.02 13.6
0.85204 0.0 8.14 0 0.5380 5.965 89.2 4.0123 4 307 21.0 392.53 13.83 19.6
1.23247 0.0 8.14 0 0.5380 6.142 91.7 3.9769 4 307 21.0 396.90 18.72 15.2
0.98843 0.0 8.14 0 0.5380 5.813 100.0 4.0952 4 307 21.0 394.54 19.88 14.5
0.75026 0.0 8.14 0 0.5380 5.924 94.1 4.3996 4 307 21.0 394.33 16.30 15.6
0.84054 0.0 8.14 0 0.5380 5.599 85.7 4.4546 4 307 21.0 303.42 16.51 13.9
0.67191 0.0 8.14 0 0.5380 5.813 90.3 4.6820 4 307 21.0 376.88 14.81 16.6
0.95577 0.0 8.14 0 0.5380 6.047 88.8 4.4534 4 307 21.0 306.38 17.28 14.8
0.77299 0.0 8.14 0 0.5380 6.495 94.4 4.4547 4 307 21.0 387.94 12.80 18.4
1.00245 0.0 8.14 0 0.5380 6.674 87.3 4.2390 4 307 21.0 380.23 11.98 21.0
1.13081 0.0 8.14 0 0.5380 5.713 94.1 4.2330 4 307 21.0 360.17 22.60 12.7
1.35472 0.0 8.14 0 0.5380 6.072 100.0 4.1750 4 307 21.0 376.73 13.04 14.5
1.38799 0.0 8.14 0 0.5380 5.950 82.0 3.9900 4 307 21.0 232.60 27.71 13.2
1.15172 0.0 8.14 0 0.5380 5.701 95.0 3.7872 4 307 21.0 358.77 18.35 13.1
1.61282 0.0 8.14 0 0.5380 6.096 96.9 3.7598 4 307 21.0 248.31 20.34 13.5
0.06417 0.0 5.96 0 0.4990 5.933 68.2 3.3603 5 279 19.2 396.90 9.68 18.9
0.09744 0.0 5.96 0 0.4990 5.841 61.4 3.3779 5 279 19.2 377.56 11.41 20.0
0.08014 0.0 5.96 0 0.4990 5.850 41.5 3.9342 5 279 19.2 396.90 8.77 21.0
0.17505 0.0 5.96 0 0.4990 5.966 30.2 3.8473 5 279 19.2 393.43 10.13 24.7
0.02763 75.0 2.95 0 0.4280 6.595 21.8 5.4011 3 252 18.3 395.63 4.32 30.8
0.03359 75.0 2.95 0 0.4280 7.024 15.8 5.4011 3 252 18.3 395.62 1.98 34.9
0.12744 0.0 6.91 0 0.4480 6.770 2.9 5.7209 3 233 17.9 385.41 4.84 26.6
0.14150 0.0 6.91 0 0.4480 6.169 6.6 5.7209 3 233 17.9 383.37 5.81 25.3
0.15936 0.0 6.91 0 0.4480 6.211 6.5 5.7209 3 233 17.9 394.46 7.44 24.7
0.12269 0.0 6.91 0 0.4480 6.069 40.0 5.7209 3 233 17.9 389.39 9.55 21.2
0.17142 0.0 6.91 0 0.4480 5.682 33.8 5.1004 3 233 17.9 396.90 10.21 19.3
0.18836 0.0 6.91 0 0.4480 5.786 33.3 5.1004 3 233 17.9 396.90 14.15 20.0
0.22927 0.0 6.91 0 0.4480 6.030 85.5 5.6894 3 233 17.9 392.74 18.80 16.6
0.25387 0.0 6.91 0 0.4480 5.399 95.3 5.8700 3 233 17.9 396.90 30.81 14.4
0.21977 0.0 6.91 0 0.4480 5.602 62.0 6.0877 3 233 17.9 396.90 16.20 19.4
0.08873 21.0 5.64 0 0.4390 5.963 45.7 6.8147 4 243 16.8 395.56 13.45 19.7
0.04337 21.0 5.64 0 0.4390 6.115 63.0 6.8147 4 243 16.8 393.97 9.43 20.5
0.05360 21.0 5.64 0 0.4390 6.511 21.1 6.8147 4 243 16.8 396.90 5.28 25.0
0.04981 21.0 5.64 0 0.4390 5.998 21.4 6.8147 4 243 16.8 396.90 8.43 23.4
0.01360 75.0 4.00 0 0.4100 5.888 47.6 7.3197 3 469 21.1 396.90 14.80 18.9
0.01311 90.0 1.22 0 0.4030 7.249 21.9 8.6966 5 226 17.9 395.93 4.81 35.4
0.02055 85.0 0.74 0 0.4100 6.383 35.7 9.1876 2 313 17.3 396.90 5.77 24.7
0.01432 100.0 1.32 0 0.4110 6.816 40.5 8.3248 5 256 15.1 392.90 3.95 31.6
0.15445 25.0 5.13 0 0.4530 6.145 29.2 7.8148 8 284 19.7 390.68 6.86 23.3
0.10328 25.0 5.13 0 0.4530 5.927 47.2 6.9320 8 284 19.7 396.90 9.22 19.6
0.14932 25.0 5.13 0 0.4530 5.741 66.2 7.2254 8 284 19.7 395.11 13.15 18.7
0.17171 25.0 5.13 0 0.4530 5.966 93.4 6.8185 8 284 19.7 378.08 14.44 16.0
0.11027 25.0 5.13 0 0.4530 6.456 67.8 7.2255 8 284 19.7 396.90 6.73 22.2
0.12650 25.0 5.13 0 0.4530 6.762 43.4 7.9809 8 284 19.7 395.58 9.50 25.0
0.01951 17.5 1.38 0 0.4161 7.104 59.5 9.2229 3 216 18.6 393.24 8.05 33.0
0.03584 80.0 3.37 0 0.3980 6.290 17.8 6.6115 4 337 16.1 396.90 4.67 23.5
0.04379 80.0 3.37 0 0.3980 5.787 31.1 6.6115 4 337 16.1 396.90 10.24 19.4
0.05789 12.5 6.07 0 0.4090 5.878 21.4 6.4980 4 345 18.9 396.21 8.10 22.0
0.13554 12.5 6.07 0 0.4090 5.594 36.8 6.4980 4 345 18.9 396.90 13.09 17.4
0.12816 12.5 6.07 0 0.4090 5.885 33.0 6.4980 4 345 18.9 396.90 8.79 20.9
0.08826 0.0 10.81 0 0.4130 6.417 6.6 5.2873 4 305 19.2 383.73 6.72 24.2
0.15876 0.0 10.81 0 0.4130 5.961 17.5 5.2873 4 305 19.2 376.94 9.88 21.7
0.09164 0.0 10.81 0 0.4130 6.065 7.8 5.2873 4 305 19.2 390.91 5.52 22.8
0.19539 0.0 10.81 0 0.4130 6.245 6.2 5.2873 4 305 19.2 377.17 7.54 23.4
0.07896 0.0 12.83 0 0.4370 6.273 6.0 4.2515 5 398 18.7 394.92 6.78 24.1
0.09512 0.0 12.83 0 0.4370 6.286 45.0 4.5026 5 398 18.7 383.23 8.94 21.4
0.10153 0.0 12.83 0 0.4370 6.279 74.5 4.0522 5 398 18.7 373.66 11.97 20.0
0.08707 0.0 12.83 0 0.4370 6.140 45.8 4.0905 5 398 18.7 386.96 10.27 20.8
0.05646 0.0 12.83 0 0.4370 6.232 53.7 5.0141 5 398 18.7 386.40 12.34 21.2
0.08387 0.0 12.83 0 0.4370 5.874 36.6 4.5026 5 398 18.7 396.06 9.10 20.3
0.04113 25.0 4.86 0 0.4260 6.727 33.5 5.4007 4 281 19.0 396.90 5.29 28.0
0.04462 25.0 4.86 0 0.4260 6.619 70.4 5.4007 4 281 19.0 395.63 7.22 23.9
0.03659 25.0 4.86 0 0.4260 6.302 32.2 5.4007 4 281 19.0 396.90 6.72 24.8
0.03551 25.0 4.86 0 0.4260 6.167 46.7 5.4007 4 281 19.0 390.64 7.51 22.9
0.05059 0.0 4.49 0 0.4490 6.389 48.0 4.7794 3 247 18.5 396.90 9.62 23.9
0.05735 0.0 4.49 0 0.4490 6.630 56.1 4.4377 3 247 18.5 392.30 6.53 26.6
0.05188 0.0 4.49 0 0.4490 6.015 45.1 4.4272 3 247 18.5 395.99 12.86 22.5
0.07151 0.0 4.49 0 0.4490 6.121 56.8 3.7476 3 247 18.5 395.15 8.44 22.2
0.05660 0.0 3.41 0 0.4890 7.007 86.3 3.4217 2 270 17.8 396.90 5.50 23.6
0.05302 0.0 3.41 0 0.4890 7.079 63.1 3.4145 2 270 17.8 396.06 5.70 28.7
0.04684 0.0 3.41 0 0.4890 6.417 66.1 3.0923 2 270 17.8 392.18 8.81 22.6
0.03932 0.0 3.41 0 0.4890 6.405 73.9 3.0921 2 270 17.8 393.55 8.20 22.0
0.04203 28.0 15.04 0 0.4640 6.442 53.6 3.6659 4 270 18.2 395.01 8.16 22.9
0.02875 28.0 15.04 0 0.4640 6.211 28.9 3.6659 4 270 18.2 396.33 6.21 25.0
0.04294 28.0 15.04 0 0.4640 6.249 77.3 3.6150 4 270 18.2 396.90 10.59 20.6
0.12204 0.0 2.89 0 0.4450 6.625 57.8 3.4952 2 276 18.0 357.98 6.65 28.4
0.11504 0.0 2.89 0 0.4450 6.163 69.6 3.4952 2 276 18.0 391.83 11.34 21.4
0.12083 0.0 2.89 0 0.4450 8.069 76.0 3.4952 2 276 18.0 396.90 4.21 38.7
0.08187 0.0 2.89 0 0.4450 7.820 36.9 3.4952 2 276 18.0 393.53 3.57 43.8
0.06860 0.0 2.89 0 0.4450 7.416 62.5 3.4952 2 276 18.0 396.90 6.19 33.2
0.14866 0.0 8.56 0 0.5200 6.727 79.9 2.7778 5 384 20.9 394.76 9.42 27.5
0.11432 0.0 8.56 0 0.5200 6.781 71.3 2.8561 5 384 20.9 395.58 7.67 26.5
0.22876 0.0 8.56 0 0.5200 6.405 85.4 2.7147 5 384 20.9 70.80 10.63 18.6
0.21161 0.0 8.56 0 0.5200 6.137 87.4 2.7147 5 384 20.9 394.47 13.44 19.3
0.13960 0.0 8.56 0 0.5200 6.167 90.0 2.4210 5 384 20.9 392.69 12.33 20.1
0.13262 0.0 8.56 0 0.5200 5.851 96.7 2.1069 5 384 20.9 394.05 16.47 19.5
0.17120 0.0 8.56 0 0.5200 5.836 91.9 2.2110 5 384 20.9 395.67 18.66 19.5
0.13117 0.0 8.56 0 0.5200 6.127 85.2 2.1224 5 384 20.9 387.69 14.09 20.4
0.12802 0.0 8.56 0 0.5200 6.474 97.1 2.4329 5 384 20.9 395.24 12.27 19.8
0.26363 0.0 8.56 0 0.5200 6.229 91.2 2.5451 5 384 20.9 391.23 15.55 19.4
0.10793 0.0 8.56 0 0.5200 6.195 54.4 2.7778 5 384 20.9 393.49 13.00 21.7
0.10084 0.0 10.01 0 0.5470 6.715 81.6 2.6775 6 432 17.8 395.59 10.16 22.8
0.12329 0.0 10.01 0 0.5470 5.913 92.9 2.3534 6 432 17.8 394.95 16.21 18.8
0.22212 0.0 10.01 0 0.5470 6.092 95.4 2.5480 6 432 17.8 396.90 17.09 18.7
0.14231 0.0 10.01 0 0.5470 6.254 84.2 2.2565 6 432 17.8 388.74 10.45 18.5
0.17134 0.0 10.01 0 0.5470 5.928 88.2 2.4631 6 432 17.8 344.91 15.76 18.3
0.13158 0.0 10.01 0 0.5470 6.176 72.5 2.7301 6 432 17.8 393.30 12.04 21.2
0.15098 0.0 10.01 0 0.5470 6.021 82.6 2.7474 6 432 17.8 394.51 10.30 19.2
0.13058 0.0 10.01 0 0.5470 5.872 73.1 2.4775 6 432 17.8 338.63 15.37 20.4
0.14476 0.0 10.01 0 0.5470 5.731 65.2 2.7592 6 432 17.8 391.50 13.61 19.3
0.06899 0.0 25.65 0 0.5810 5.870 69.7 2.2577 2 188 19.1 389.15 14.37 22.0
0.07165 0.0 25.65 0 0.5810 6.004 84.1 2.1974 2 188 19.1 377.67 14.27 20.3
0.09299 0.0 25.65 0 0.5810 5.961 92.9 2.0869 2 188 19.1 378.09 17.93 20.5
0.15038 0.0 25.65 0 0.5810 5.856 97.0 1.9444 2 188 19.1 370.31 25.41 17.3
0.09849 0.0 25.65 0 0.5810 5.879 95.8 2.0063 2 188 19.1 379.38 17.58 18.8
0.16902 0.0 25.65 0 0.5810 5.986 88.4 1.9929 2 188 19.1 385.02 14.81 21.4
0.38735 0.0 25.65 0 0.5810 5.613 95.6 1.7572 2 188 19.1 359.29 27.26 15.7
0.25915 0.0 21.89 0 0.6240 5.693 96.0 1.7883 4 437 21.2 392.11 17.19 16.2
0.32543 0.0 21.89 0 0.6240 6.431 98.8 1.8125 4 437 21.2 396.90 15.39 18.0
0.88125 0.0 21.89 0 0.6240 5.637 94.7 1.9799 4 437 21.2 396.90 18.34 14.3
0.34006 0.0 21.89 0 0.6240 6.458 98.9 2.1185 4 437 21.2 395.04 12.60 19.2
1.19294 0.0 21.89 0 0.6240 6.326 97.7 2.2710 4 437 21.2 396.90 12.26 19.6
0.59005 0.0 21.89 0 0.6240 6.372 97.9 2.3274 4 437 21.2 385.76 11.12 23.0
0.32982 0.0 21.89 0 0.6240 5.822 95.4 2.4699 4 437 21.2 388.69 15.03 18.4
0.97617 0.0 21.89 0 0.6240 5.757 98.4 2.3460 4 437 21.2 262.76 17.31 15.6
0.55778 0.0 21.89 0 0.6240 6.335 98.2 2.1107 4 437 21.2 394.67 16.96 18.1
0.32264 0.0 21.89 0 0.6240 5.942 93.5 1.9669 4 437 21.2 378.25 16.90 17.4
0.35233 0.0 21.89 0 0.6240 6.454 98.4 1.8498 4 437 21.2 394.08 14.59 17.1
0.24980 0.0 21.89 0 0.6240 5.857 98.2 1.6686 4 437 21.2 392.04 21.32 13.3
0.54452 0.0 21.89 0 0.6240 6.151 97.9 1.6687 4 437 21.2 396.90 18.46 17.8
0.29090 0.0 21.89 0 0.6240 6.174 93.6 1.6119 4 437 21.2 388.08 24.16 14.0
1.62864 0.0 21.89 0 0.6240 5.019 100.0 1.4394 4 437 21.2 396.90 34.41 14.4
3.32105 0.0 19.58 1 0.8710 5.403 100.0 1.3216 5 403 14.7 396.90 26.82 13.4
4.09740 0.0 19.58 0 0.8710 5.468 100.0 1.4118 5 403 14.7 396.90 26.42 15.6
2.77974 0.0 19.58 0 0.8710 4.903 97.8 1.3459 5 403 14.7 396.90 29.29 11.8
2.37934 0.0 19.58 0 0.8710 6.130 100.0 1.4191 5 403 14.7 172.91 27.80 13.8
2.15505 0.0 19.58 0 0.8710 5.628 100.0 1.5166 5 403 14.7 169.27 16.65 15.6
2.36862 0.0 19.58 0 0.8710 4.926 95.7 1.4608 5 403 14.7 391.71 29.53 14.6
2.33099 0.0 19.58 0 0.8710 5.186 93.8 1.5296 5 403 14.7 356.99 28.32 17.8
2.73397 0.0 19.58 0 0.8710 5.597 94.9 1.5257 5 403 14.7 351.85 21.45 15.4
1.65660 0.0 19.58 0 0.8710 6.122 97.3 1.6180 5 403 14.7 372.80 14.10 21.5
1.49632 0.0 19.58 0 0.8710 5.404 100.0 1.5916 5 403 14.7 341.60 13.28 19.6
1.12658 0.0 19.58 1 0.8710 5.012 88.0 1.6102 5 403 14.7 343.28 12.12 15.3
2.14918 0.0 19.58 0 0.8710 5.709 98.5 1.6232 5 403 14.7 261.95 15.79 19.4
1.41385 0.0 19.58 1 0.8710 6.129 96.0 1.7494 5 403 14.7 321.02 15.12 17.0
3.53501 0.0 19.58 1 0.8710 6.152 82.6 1.7455 5 403 14.7 88.01 15.02 15.6
2.44668 0.0 19.58 0 0.8710 5.272 94.0 1.7364 5 403 14.7 88.63 16.14 13.1
1.22358 0.0 19.58 0 0.6050 6.943 97.4 1.8773 5 403 14.7 363.43 4.59 41.3
1.34284 0.0 19.58 0 0.6050 6.066 100.0 1.7573 5 403 14.7 353.89 6.43 24.3
1.42502 0.0 19.58 0 0.8710 6.510 100.0 1.7659 5 403 14.7 364.31 7.39 23.3
1.27346 0.0 19.58 1 0.6050 6.250 92.6 1.7984 5 403 14.7 338.92 5.50 27.0
1.46336 0.0 19.58 0 0.6050 7.489 90.8 1.9709 5 403 14.7 374.43 1.73 50.0
1.83377 0.0 19.58 1 0.6050 7.802 98.2 2.0407 5 403 14.7 389.61 1.92 50.0
1.51902 0.0 19.58 1 0.6050 8.375 93.9 2.1620 5 403 14.7 388.45 3.32 50.0
2.24236 0.0 19.58 0 0.6050 5.854 91.8 2.4220 5 403 14.7 395.11 11.64 22.7
2.92400 0.0 19.58 0 0.6050 6.101 93.0 2.2834 5 403 14.7 240.16 9.81 25.0
2.01019 0.0 19.58 0 0.6050 7.929 96.2 2.0459 5 403 14.7 369.30 3.70 50.0
1.80028 0.0 19.58 0 0.6050 5.877 79.2 2.4259 5 403 14.7 227.61 12.14 23.8
2.30040 0.0 19.58 0 0.6050 6.319 96.1 2.1000 5 403 14.7 297.09 11.10 23.8
2.44953 0.0 19.58 0 0.6050 6.402 95.2 2.2625 5 403 14.7 330.04 11.32 22.3
1.20742 0.0 19.58 0 0.6050 5.875 94.6 2.4259 5 403 14.7 292.29 14.43 17.4
2.31390 0.0 19.58 0 0.6050 5.880 97.3 2.3887 5 403 14.7 348.13 12.03 19.1
0.13914 0.0 4.05 0 0.5100 5.572 88.5 2.5961 5 296 16.6 396.90 14.69 23.1
0.09178 0.0 4.05 0 0.5100 6.416 84.1 2.6463 5 296 16.6 395.50 9.04 23.6
0.08447 0.0 4.05 0 0.5100 5.859 68.7 2.7019 5 296 16.6 393.23 9.64 22.6
0.06664 0.0 4.05 0 0.5100 6.546 33.1 3.1323 5 296 16.6 390.96 5.33 29.4
0.07022 0.0 4.05 0 0.5100 6.020 47.2 3.5549 5 296 16.6 393.23 10.11 23.2
0.05425 0.0 4.05 0 0.5100 6.315 73.4 3.3175 5 296 16.6 395.60 6.29 24.6
0.06642 0.0 4.05 0 0.5100 6.860 74.4 2.9153 5 296 16.6 391.27 6.92 29.9
0.05780 0.0 2.46 0 0.4880 6.980 58.4 2.8290 3 193 17.8 396.90 5.04 37.2
0.06588 0.0 2.46 0 0.4880 7.765 83.3 2.7410 3 193 17.8 395.56 7.56 39.8
0.06888 0.0 2.46 0 0.4880 6.144 62.2 2.5979 3 193 17.8 396.90 9.45 36.2
0.09103 0.0 2.46 0 0.4880 7.155 92.2 2.7006 3 193 17.8 394.12 4.82 37.9
0.10008 0.0 2.46 0 0.4880 6.563 95.6 2.8470 3 193 17.8 396.90 5.68 32.5
0.08308 0.0 2.46 0 0.4880 5.604 89.8 2.9879 3 193 17.8 391.00 13.98 26.4
0.06047 0.0 2.46 0 0.4880 6.153 68.8 3.2797 3 193 17.8 387.11 13.15 29.6
0.05602 0.0 2.46 0 0.4880 7.831 53.6 3.1992 3 193 17.8 392.63 4.45 50.0
0.07875 45.0 3.44 0 0.4370 6.782 41.1 3.7886 5 398 15.2 393.87 6.68 32.0
0.12579 45.0 3.44 0 0.4370 6.556 29.1 4.5667 5 398 15.2 382.84 4.56 29.8
0.08370 45.0 3.44 0 0.4370 7.185 38.9 4.5667 5 398 15.2 396.90 5.39 34.9
0.09068 45.0 3.44 0 0.4370 6.951 21.5 6.4798 5 398 15.2 377.68 5.10 37.0
0.06911 45.0 3.44 0 0.4370 6.739 30.8 6.4798 5 398 15.2 389.71 4.69 30.5
0.08664 45.0 3.44 0 0.4370 7.178 26.3 6.4798 5 398 15.2 390.49 2.87 36.4
0.02187 60.0 2.93 0 0.4010 6.800 9.9 6.2196 1 265 15.6 393.37 5.03 31.1
0.01439 60.0 2.93 0 0.4010 6.604 18.8 6.2196 1 265 15.6 376.70 4.38 29.1
0.01381 80.0 0.46 0 0.4220 7.875 32.0 5.6484 4 255 14.4 394.23 2.97 50.0
0.04011 80.0 1.52 0 0.4040 7.287 34.1 7.3090 2 329 12.6 396.90 4.08 33.3
0.04666 80.0 1.52 0 0.4040 7.107 36.6 7.3090 2 329 12.6 354.31 8.61 30.3
0.03768 80.0 1.52 0 0.4040 7.274 38.3 7.3090 2 329 12.6 392.20 6.62 34.6
0.03150 95.0 1.47 0 0.4030 6.975 15.3 7.6534 3 402 17.0 396.90 4.56 34.9
0.01778 95.0 1.47 0 0.4030 7.135 13.9 7.6534 3 402 17.0 384.30 4.45 32.9
0.03445 82.5 2.03 0 0.4150 6.162 38.4 6.2700 2 348 14.7 393.77 7.43 24.1
0.02177 82.5 2.03 0 0.4150 7.610 15.7 6.2700 2 348 14.7 395.38 3.11 42.3
0.03510 95.0 2.68 0 0.4161 7.853 33.2 5.1180 4 224 14.7 392.78 3.81 48.5
0.02009 95.0 2.68 0 0.4161 8.034 31.9 5.1180 4 224 14.7 390.55 2.88 50.0
0.13642 0.0 10.59 0 0.4890 5.891 22.3 3.9454 4 277 18.6 396.90 10.87 22.6
0.22969 0.0 10.59 0 0.4890 6.326 52.5 4.3549 4 277 18.6 394.87 10.97 24.4
0.25199 0.0 10.59 0 0.4890 5.783 72.7 4.3549 4 277 18.6 389.43 18.06 22.5
0.13587 0.0 10.59 1 0.4890 6.064 59.1 4.2392 4 277 18.6 381.32 14.66 24.4
0.43571 0.0 10.59 1 0.4890 5.344 100.0 3.8750 4 277 18.6 396.90 23.09 20.0
0.17446 0.0 10.59 1 0.4890 5.960 92.1 3.8771 4 277 18.6 393.25 17.27 21.7
0.37578 0.0 10.59 1 0.4890 5.404 88.6 3.6650 4 277 18.6 395.24 23.98 19.3
0.21719 0.0 10.59 1 0.4890 5.807 53.8 3.6526 4 277 18.6 390.94 16.03 22.4
0.14052 0.0 10.59 0 0.4890 6.375 32.3 3.9454 4 277 18.6 385.81 9.38 28.1
0.28955 0.0 10.59 0 0.4890 5.412 9.8 3.5875 4 277 18.6 348.93 29.55 23.7
0.19802 0.0 10.59 0 0.4890 6.182 42.4 3.9454 4 277 18.6 393.63 9.47 25.0
0.04560 0.0 13.89 1 0.5500 5.888 56.0 3.1121 5 276 16.4 392.80 13.51 23.3
0.07013 0.0 13.89 0 0.5500 6.642 85.1 3.4211 5 276 16.4 392.78 9.69 28.7
0.11069 0.0 13.89 1 0.5500 5.951 93.8 2.8893 5 276 16.4 396.90 17.92 21.5
0.11425 0.0 13.89 1 0.5500 6.373 92.4 3.3633 5 276 16.4 393.74 10.50 23.0
0.35809 0.0 6.20 1 0.5070 6.951 88.5 2.8617 8 307 17.4 391.70 9.71 26.7
0.40771 0.0 6.20 1 0.5070 6.164 91.3 3.0480 8 307 17.4 395.24 21.46 21.7
0.62356 0.0 6.20 1 0.5070 6.879 77.7 3.2721 8 307 17.4 390.39 9.93 27.5
0.61470 0.0 6.20 0 0.5070 6.618 80.8 3.2721 8 307 17.4 396.90 7.60 30.1
0.31533 0.0 6.20 0 0.5040 8.266 78.3 2.8944 8 307 17.4 385.05 4.14 44.8
0.52693 0.0 6.20 0 0.5040 8.725 83.0 2.8944 8 307 17.4 382.00 4.63 50.0
0.38214 0.0 6.20 0 0.5040 8.040 86.5 3.2157 8 307 17.4 387.38 3.13 37.6
0.41238 0.0 6.20 0 0.5040 7.163 79.9 3.2157 8 307 17.4 372.08 6.36 31.6
0.29819 0.0 6.20 0 0.5040 7.686 17.0 3.3751 8 307 17.4 377.51 3.92 46.7
0.44178 0.0 6.20 0 0.5040 6.552 21.4 3.3751 8 307 17.4 380.34 3.76 31.5
0.53700 0.0 6.20 0 0.5040 5.981 68.1 3.6715 8 307 17.4 378.35 11.65 24.3
0.46296 0.0 6.20 0 0.5040 7.412 76.9 3.6715 8 307 17.4 376.14 5.25 31.7
0.57529 0.0 6.20 0 0.5070 8.337 73.3 3.8384 8 307 17.4 385.91 2.47 41.7
0.33147 0.0 6.20 0 0.5070 8.247 70.4 3.6519 8 307 17.4 378.95 3.95 48.3
0.44791 0.0 6.20 1 0.5070 6.726 66.5 3.6519 8 307 17.4 360.20 8.05 29.0
0.33045 0.0 6.20 0 0.5070 6.086 61.5 3.6519 8 307 17.4 376.75 10.88 24.0
0.52058 0.0 6.20 1 0.5070 6.631 76.5 4.1480 8 307 17.4 388.45 9.54 25.1
0.51183 0.0 6.20 0 0.5070 7.358 71.6 4.1480 8 307 17.4 390.07 4.73 31.5
0.08244 30.0 4.93 0 0.4280 6.481 18.5 6.1899 6 300 16.6 379.41 6.36 23.7
0.09252 30.0 4.93 0 0.4280 6.606 42.2 6.1899 6 300 16.6 383.78 7.37 23.3
0.11329 30.0 4.93 0 0.4280 6.897 54.3 6.3361 6 300 16.6 391.25 11.38 22.0
0.10612 30.0 4.93 0 0.4280 6.095 65.1 6.3361 6 300 16.6 394.62 12.40 20.1
0.10290 30.0 4.93 0 0.4280 6.358 52.9 7.0355 6 300 16.6 372.75 11.22 22.2
0.12757 30.0 4.93 0 0.4280 6.393 7.8 7.0355 6 300 16.6 374.71 5.19 23.7
0.20608 22.0 5.86 0 0.4310 5.593 76.5 7.9549 7 330 19.1 372.49 12.50 17.6
0.19133 22.0 5.86 0 0.4310 5.605 70.2 7.9549 7 330 19.1 389.13 18.46 18.5
0.33983 22.0 5.86 0 0.4310 6.108 34.9 8.0555 7 330 19.1 390.18 9.16 24.3
0.19657 22.0 5.86 0 0.4310 6.226 79.2 8.0555 7 330 19.1 376.14 10.15 20.5
0.16439 22.0 5.86 0 0.4310 6.433 49.1 7.8265 7 330 19.1 374.71 9.52 24.5
0.19073 22.0 5.86 0 0.4310 6.718 17.5 7.8265 7 330 19.1 393.74 6.56 26.2
0.14030 22.0 5.86 0 0.4310 6.487 13.0 7.3967 7 330 19.1 396.28 5.90 24.4
0.21409 22.0 5.86 0 0.4310 6.438 8.9 7.3967 7 330 19.1 377.07 3.59 24.8
0.08221 22.0 5.86 0 0.4310 6.957 6.8 8.9067 7 330 19.1 386.09 3.53 29.6
0.36894 22.0 5.86 0 0.4310 8.259 8.4 8.9067 7 330 19.1 396.90 3.54 42.8
0.04819 80.0 3.64 0 0.3920 6.108 32.0 9.2203 1 315 16.4 392.89 6.57 21.9
0.03548 80.0 3.64 0 0.3920 5.876 19.1 9.2203 1 315 16.4 395.18 9.25 20.9
0.01538 90.0 3.75 0 0.3940 7.454 34.2 6.3361 3 244 15.9 386.34 3.11 44.0
0.61154 20.0 3.97 0 0.6470 8.704 86.9 1.8010 5 264 13.0 389.70 5.12 50.0
0.66351 20.0 3.97 0 0.6470 7.333 100.0 1.8946 5 264 13.0 383.29 7.79 36.0
0.65665 20.0 3.97 0 0.6470 6.842 100.0 2.0107 5 264 13.0 391.93 6.90 30.1
0.54011 20.0 3.97 0 0.6470 7.203 81.8 2.1121 5 264 13.0 392.80 9.59 33.8
0.53412 20.0 3.97 0 0.6470 7.520 89.4 2.1398 5 264 13.0 388.37 7.26 43.1
0.52014 20.0 3.97 0 0.6470 8.398 91.5 2.2885 5 264 13.0 386.86 5.91 48.8
0.82526 20.0 3.97 0 0.6470 7.327 94.5 2.0788 5 264 13.0 393.42 11.25 31.0
0.55007 20.0 3.97 0 0.6470 7.206 91.6 1.9301 5 264 13.0 387.89 8.10 36.5
0.76162 20.0 3.97 0 0.6470 5.560 62.8 1.9865 5 264 13.0 392.40 10.45 22.8
0.78570 20.0 3.97 0 0.6470 7.014 84.6 2.1329 5 264 13.0 384.07 14.79 30.7
0.57834 20.0 3.97 0 0.5750 8.297 67.0 2.4216 5 264 13.0 384.54 7.44 50.0
0.54050 20.0 3.97 0 0.5750 7.470 52.6 2.8720 5 264 13.0 390.30 3.16 43.5
0.09065 20.0 6.96 1 0.4640 5.920 61.5 3.9175 3 223 18.6 391.34 13.65 20.7
0.29916 20.0 6.96 0 0.4640 5.856 42.1 4.4290 3 223 18.6 388.65 13.00 21.1
0.16211 20.0 6.96 0 0.4640 6.240 16.3 4.4290 3 223 18.6 396.90 6.59 25.2
0.11460 20.0 6.96 0 0.4640 6.538 58.7 3.9175 3 223 18.6 394.96 7.73 24.4
0.22188 20.0 6.96 1 0.4640 7.691 51.8 4.3665 3 223 18.6 390.77 6.58 35.2
0.05644 40.0 6.41 1 0.4470 6.758 32.9 4.0776 4 254 17.6 396.90 3.53 32.4
0.09604 40.0 6.41 0 0.4470 6.854 42.8 4.2673 4 254 17.6 396.90 2.98 32.0
0.10469 40.0 6.41 1 0.4470 7.267 49.0 4.7872 4 254 17.6 389.25 6.05 33.2
0.06127 40.0 6.41 1 0.4470 6.826 27.6 4.8628 4 254 17.6 393.45 4.16 33.1
0.07978 40.0 6.41 0 0.4470 6.482 32.1 4.1403 4 254 17.6 396.90 7.19 29.1
0.21038 20.0 3.33 0 0.4429 6.812 32.2 4.1007 5 216 14.9 396.90 4.85 35.1
0.03578 20.0 3.33 0 0.4429 7.820 64.5 4.6947 5 216 14.9 387.31 3.76 45.4
0.03705 20.0 3.33 0 0.4429 6.968 37.2 5.2447 5 216 14.9 392.23 4.59 35.4
0.06129 20.0 3.33 1 0.4429 7.645 49.7 5.2119 5 216 14.9 377.07 3.01 46.0
0.01501 90.0 1.21 1 0.4010 7.923 24.8 5.8850 1 198 13.6 395.52 3.16 50.0
0.00906 90.0 2.97 0 0.4000 7.088 20.8 7.3073 1 285 15.3 394.72 7.85 32.2
0.01096 55.0 2.25 0 0.3890 6.453 31.9 7.3073 1 300 15.3 394.72 8.23 22.0
0.01965 80.0 1.76 0 0.3850 6.230 31.5 9.0892 1 241 18.2 341.60 12.93 20.1
0.03871 52.5 5.32 0 0.4050 6.209 31.3 7.3172 6 293 16.6 396.90 7.14 23.2
0.04590 52.5 5.32 0 0.4050 6.315 45.6 7.3172 6 293 16.6 396.90 7.60 22.3
0.04297 52.5 5.32 0 0.4050 6.565 22.9 7.3172 6 293 16.6 371.72 9.51 24.8
0.03502 80.0 4.95 0 0.4110 6.861 27.9 5.1167 4 245 19.2 396.90 3.33 28.5
0.07886 80.0 4.95 0 0.4110 7.148 27.7 5.1167 4 245 19.2 396.90 3.56 37.3
0.03615 80.0 4.95 0 0.4110 6.630 23.4 5.1167 4 245 19.2 396.90 4.70 27.9
0.08265 0.0 13.92 0 0.4370 6.127 18.4 5.5027 4 289 16.0 396.90 8.58 23.9
0.08199 0.0 13.92 0 0.4370 6.009 42.3 5.5027 4 289 16.0 396.90 10.40 21.7
0.12932 0.0 13.92 0 0.4370 6.678 31.1 5.9604 4 289 16.0 396.90 6.27 28.6
0.05372 0.0 13.92 0 0.4370 6.549 51.0 5.9604 4 289 16.0 392.85 7.39 27.1
0.14103 0.0 13.92 0 0.4370 5.790 58.0 6.3200 4 289 16.0 396.90 15.84 20.3
0.06466 70.0 2.24 0 0.4000 6.345 20.1 7.8278 5 358 14.8 368.24 4.97 22.5
0.05561 70.0 2.24 0 0.4000 7.041 10.0 7.8278 5 358 14.8 371.58 4.74 29.0
0.04417 70.0 2.24 0 0.4000 6.871 47.4 7.8278 5 358 14.8 390.86 6.07 24.8
0.03537 34.0 6.09 0 0.4330 6.590 40.4 5.4917 7 329 16.1 395.75 9.50 22.0
0.09266 34.0 6.09 0 0.4330 6.495 18.4 5.4917 7 329 16.1 383.61 8.67 26.4
0.10000 34.0 6.09 0 0.4330 6.982 17.7 5.4917 7 329 16.1 390.43 4.86 33.1
0.05515 33.0 2.18 0 0.4720 7.236 41.1 4.0220 7 222 18.4 393.68 6.93 36.1
0.05479 33.0 2.18 0 0.4720 6.616 58.1 3.3700 7 222 18.4 393.36 8.93 28.4
0.07503 33.0 2.18 0 0.4720 7.420 71.9 3.0992 7 222 18.4 396.90 6.47 33.4
0.04932 33.0 2.18 0 0.4720 6.849 70.3 3.1827 7 222 18.4 396.90 7.53 28.2
0.49298 0.0 9.90 0 0.5440 6.635 82.5 3.3175 4 304 18.4 396.90 4.54 22.8
0.34940 0.0 9.90 0 0.5440 5.972 76.7 3.1025 4 304 18.4 396.24 9.97 20.3
2.63548 0.0 9.90 0 0.5440 4.973 37.8 2.5194 4 304 18.4 350.45 12.64 16.1
0.79041 0.0 9.90 0 0.5440 6.122 52.8 2.6403 4 304 18.4 396.90 5.98 22.1
0.26169 0.0 9.90 0 0.5440 6.023 90.4 2.8340 4 304 18.4 396.30 11.72 19.4
0.26938 0.0 9.90 0 0.5440 6.266 82.8 3.2628 4 304 18.4 393.39 7.90 21.6
0.36920 0.0 9.90 0 0.5440 6.567 87.3 3.6023 4 304 18.4 395.69 9.28 23.8
0.25356 0.0 9.90 0 0.5440 5.705 77.7 3.9450 4 304 18.4 396.42 11.50 16.2
0.31827 0.0 9.90 0 0.5440 5.914 83.2 3.9986 4 304 18.4 390.70 18.33 17.8
0.24522 0.0 9.90 0 0.5440 5.782 71.7 4.0317 4 304 18.4 396.90 15.94 19.8
0.40202 0.0 9.90 0 0.5440 6.382 67.2 3.5325 4 304 18.4 395.21 10.36 23.1
0.47547 0.0 9.90 0 0.5440 6.113 58.8 4.0019 4 304 18.4 396.23 12.73 21.0
0.16760 0.0 7.38 0 0.4930 6.426 52.3 4.5404 5 287 19.6 396.90 7.20 23.8
0.18159 0.0 7.38 0 0.4930 6.376 54.3 4.5404 5 287 19.6 396.90 6.87 23.1
0.35114 0.0 7.38 0 0.4930 6.041 49.9 4.7211 5 287 19.6 396.90 7.70 20.4
0.28392 0.0 7.38 0 0.4930 5.708 74.3 4.7211 5 287 19.6 391.13 11.74 18.5
0.34109 0.0 7.38 0 0.4930 6.415 40.1 4.7211 5 287 19.6 396.90 6.12 25.0
0.19186 0.0 7.38 0 0.4930 6.431 14.7 5.4159 5 287 19.6 393.68 5.08 24.6
0.30347 0.0 7.38 0 0.4930 6.312 28.9 5.4159 5 287 19.6 396.90 6.15 23.0
0.24103 0.0 7.38 0 0.4930 6.083 43.7 5.4159 5 287 19.6 396.90 12.79 22.2
0.06617 0.0 3.24 0 0.4600 5.868 25.8 5.2146 4 430 16.9 382.44 9.97 19.3
0.06724 0.0 3.24 0 0.4600 6.333 17.2 5.2146 4 430 16.9 375.21 7.34 22.6
0.04544 0.0 3.24 0 0.4600 6.144 32.2 5.8736 4 430 16.9 368.57 9.09 19.8
0.05023 35.0 6.06 0 0.4379 5.706 28.4 6.6407 1 304 16.9 394.02 12.43 17.1
0.03466 35.0 6.06 0 0.4379 6.031 23.3 6.6407 1 304 16.9 362.25 7.83 19.4
0.05083 0.0 5.19 0 0.5150 6.316 38.1 6.4584 5 224 20.2 389.71 5.68 22.2
0.03738 0.0 5.19 0 0.5150 6.310 38.5 6.4584 5 224 20.2 389.40 6.75 20.7
0.03961 0.0 5.19 0 0.5150 6.037 34.5 5.9853 5 224 20.2 396.90 8.01 21.1
0.03427 0.0 5.19 0 0.5150 5.869 46.3 5.2311 5 224 20.2 396.90 9.80 19.5
0.03041 0.0 5.19 0 0.5150 5.895 59.6 5.6150 5 224 20.2 394.81 10.56 18.5
0.03306 0.0 5.19 0 0.5150 6.059 37.3 4.8122 5 224 20.2 396.14 8.51 20.6
0.05497 0.0 5.19 0 0.5150 5.985 45.4 4.8122 5 224 20.2 396.90 9.74 19.0
0.06151 0.0 5.19 0 0.5150 5.968 58.5 4.8122 5 224 20.2 396.90 9.29 18.7
0.01301 35.0 1.52 0 0.4420 7.241 49.3 7.0379 1 284 15.5 394.74 5.49 32.7
0.02498 0.0 1.89 0 0.5180 6.540 59.7 6.2669 1 422 15.9 389.96 8.65 16.5
0.02543 55.0 3.78 0 0.4840 6.696 56.4 5.7321 5 370 17.6 396.90 7.18 23.9
0.03049 55.0 3.78 0 0.4840 6.874 28.1 6.4654 5 370 17.6 387.97 4.61 31.2
0.03113 0.0 4.39 0 0.4420 6.014 48.5 8.0136 3 352 18.8 385.64 10.53 17.5
0.06162 0.0 4.39 0 0.4420 5.898 52.3 8.0136 3 352 18.8 364.61 12.67 17.2
0.01870 85.0 4.15 0 0.4290 6.516 27.7 8.5353 4 351 17.9 392.43 6.36 23.1
0.01501 80.0 2.01 0 0.4350 6.635 29.7 8.3440 4 280 17.0 390.94 5.99 24.5
0.02899 40.0 1.25 0 0.4290 6.939 34.5 8.7921 1 335 19.7 389.85 5.89 26.6
0.06211 40.0 1.25 0 0.4290 6.490 44.4 8.7921 1 335 19.7 396.90 5.98 22.9
0.07950 60.0 1.69 0 0.4110 6.579 35.9 10.7103 4 411 18.3 370.78 5.49 24.1
0.07244 60.0 1.69 0 0.4110 5.884 18.5 10.7103 4 411 18.3 392.33 7.79 18.6
0.01709 90.0 2.02 0 0.4100 6.728 36.1 12.1265 5 187 17.0 384.46 4.50 30.1
0.04301 80.0 1.91 0 0.4130 5.663 21.9 10.5857 4 334 22.0 382.80 8.05 18.2
0.10659 80.0 1.91 0 0.4130 5.936 19.5 10.5857 4 334 22.0 376.04 5.57 20.6
8.98296 0.0 18.10 1 0.7700 6.212 97.4 2.1222 24 666 20.2 377.73 17.60 17.8
3.84970 0.0 18.10 1 0.7700 6.395 91.0 2.5052 24 666 20.2 391.34 13.27 21.7
5.20177 0.0 18.10 1 0.7700 6.127 83.4 2.7227 24 666 20.2 395.43 11.48 22.7
4.26131 0.0 18.10 0 0.7700 6.112 81.3 2.5091 24 666 20.2 390.74 12.67 22.6
4.54192 0.0 18.10 0 0.7700 6.398 88.0 2.5182 24 666 20.2 374.56 7.79 25.0
3.83684 0.0 18.10 0 0.7700 6.251 91.1 2.2955 24 666 20.2 350.65 14.19 19.9
3.67822 0.0 18.10 0 0.7700 5.362 96.2 2.1036 24 666 20.2 380.79 10.19 20.8
4.22239 0.0 18.10 1 0.7700 5.803 89.0 1.9047 24 666 20.2 353.04 14.64 16.8
3.47428 0.0 18.10 1 0.7180 8.780 82.9 1.9047 24 666 20.2 354.55 5.29 21.9
4.55587 0.0 18.10 0 0.7180 3.561 87.9 1.6132 24 666 20.2 354.70 7.12 27.5
3.69695 0.0 18.10 0 0.7180 4.963 91.4 1.7523 24 666 20.2 316.03 14.00 21.9
13.52220 0.0 18.10 0 0.6310 3.863 100.0 1.5106 24 666 20.2 131.42 13.33 23.1
4.89822 0.0 18.10 0 0.6310 4.970 100.0 1.3325 24 666 20.2 375.52 3.26 50.0
5.66998 0.0 18.10 1 0.6310 6.683 96.8 1.3567 24 666 20.2 375.33 3.73 50.0
6.53876 0.0 18.10 1 0.6310 7.016 97.5 1.2024 24 666 20.2 392.05 2.96 50.0
9.23230 0.0 18.10 0 0.6310 6.216 100.0 1.1691 24 666 20.2 366.15 9.53 50.0
8.26725 0.0 18.10 1 0.6680 5.875 89.6 1.1296 24 666 20.2 347.88 8.88 50.0
11.10810 0.0 18.10 0 0.6680 4.906 100.0 1.1742 24 666 20.2 396.90 34.77 13.8
18.49820 0.0 18.10 0 0.6680 4.138 100.0 1.1370 24 666 20.2 396.90 37.97 13.8
19.60910 0.0 18.10 0 0.6710 7.313 97.9 1.3163 24 666 20.2 396.90 13.44 15.0
15.28800 0.0 18.10 0 0.6710 6.649 93.3 1.3449 24 666 20.2 363.02 23.24 13.9
9.82349 0.0 18.10 0 0.6710 6.794 98.8 1.3580 24 666 20.2 396.90 21.24 13.3
23.64820 0.0 18.10 0 0.6710 6.380 96.2 1.3861 24 666 20.2 396.90 23.69 13.1
17.86670 0.0 18.10 0 0.6710 6.223 100.0 1.3861 24 666 20.2 393.74 21.78 10.2
88.97620 0.0 18.10 0 0.6710 6.968 91.9 1.4165 24 666 20.2 396.90 17.21 10.4
15.87440 0.0 18.10 0 0.6710 6.545 99.1 1.5192 24 666 20.2 396.90 21.08 10.9
9.18702 0.0 18.10 0 0.7000 5.536 100.0 1.5804 24 666 20.2 396.90 23.60 11.3
7.99248 0.0 18.10 0 0.7000 5.520 100.0 1.5331 24 666 20.2 396.90 24.56 12.3
20.08490 0.0 18.10 0 0.7000 4.368 91.2 1.4395 24 666 20.2 285.83 30.63 8.8
16.81180 0.0 18.10 0 0.7000 5.277 98.1 1.4261 24 666 20.2 396.90 30.81 7.2
24.39380 0.0 18.10 0 0.7000 4.652 100.0 1.4672 24 666 20.2 396.90 28.28 10.5
22.59710 0.0 18.10 0 0.7000 5.000 89.5 1.5184 24 666 20.2 396.90 31.99 7.4
14.33370 0.0 18.10 0 0.7000 4.880 100.0 1.5895 24 666 20.2 372.92 30.62 10.2
8.15174 0.0 18.10 0 0.7000 5.390 98.9 1.7281 24 666 20.2 396.90 20.85 11.5
6.96215 0.0 18.10 0 0.7000 5.713 97.0 1.9265 24 666 20.2 394.43 17.11 15.1
5.29305 0.0 18.10 0 0.7000 6.051 82.5 2.1678 24 666 20.2 378.38 18.76 23.2
11.57790 0.0 18.10 0 0.7000 5.036 97.0 1.7700 24 666 20.2 396.90 25.68 9.7
8.64476 0.0 18.10 0 0.6930 6.193 92.6 1.7912 24 666 20.2 396.90 15.17 13.8
13.35980 0.0 18.10 0 0.6930 5.887 94.7 1.7821 24 666 20.2 396.90 16.35 12.7
8.71675 0.0 18.10 0 0.6930 6.471 98.8 1.7257 24 666 20.2 391.98 17.12 13.1
5.87205 0.0 18.10 0 0.6930 6.405 96.0 1.6768 24 666 20.2 396.90 19.37 12.5
7.67202 0.0 18.10 0 0.6930 5.747 98.9 1.6334 24 666 20.2 393.10 19.92 8.5
38.35180 0.0 18.10 0 0.6930 5.453 100.0 1.4896 24 666 20.2 396.90 30.59 5.0
9.91655 0.0 18.10 0 0.6930 5.852 77.8 1.5004 24 666 20.2 338.16 29.97 6.3
25.04610 0.0 18.10 0 0.6930 5.987 100.0 1.5888 24 666 20.2 396.90 26.77 5.6
14.23620 0.0 18.10 0 0.6930 6.343 100.0 1.5741 24 666 20.2 396.90 20.32 7.2
9.59571 0.0 18.10 0 0.6930 6.404 100.0 1.6390 24 666 20.2 376.11 20.31 12.1
24.80170 0.0 18.10 0 0.6930 5.349 96.0 1.7028 24 666 20.2 396.90 19.77 8.3
41.52920 0.0 18.10 0 0.6930 5.531 85.4 1.6074 24 666 20.2 329.46 27.38 8.5
67.92080 0.0 18.10 0 0.6930 5.683 100.0 1.4254 24 666 20.2 384.97 22.98 5.0
20.71620 0.0 18.10 0 0.6590 4.138 100.0 1.1781 24 666 20.2 370.22 23.34 11.9
11.95110 0.0 18.10 0 0.6590 5.608 100.0 1.2852 24 666 20.2 332.09 12.13 27.9
7.40389 0.0 18.10 0 0.5970 5.617 97.9 1.4547 24 666 20.2 314.64 26.40 17.2
14.43830 0.0 18.10 0 0.5970 6.852 100.0 1.4655 24 666 20.2 179.36 19.78 27.5
51.13580 0.0 18.10 0 0.5970 5.757 100.0 1.4130 24 666 20.2 2.60 10.11 15.0
14.05070 0.0 18.10 0 0.5970 6.657 100.0 1.5275 24 666 20.2 35.05 21.22 17.2
18.81100 0.0 18.10 0 0.5970 4.628 100.0 1.5539 24 666 20.2 28.79 34.37 17.9
28.65580 0.0 18.10 0 0.5970 5.155 100.0 1.5894 24 666 20.2 210.97 20.08 16.3
45.74610 0.0 18.10 0 0.6930 4.519 100.0 1.6582 24 666 20.2 88.27 36.98 7.0
18.08460 0.0 18.10 0 0.6790 6.434 100.0 1.8347 24 666 20.2 27.25 29.05 7.2
10.83420 0.0 18.10 0 0.6790 6.782 90.8 1.8195 24 666 20.2 21.57 25.79 7.5
25.94060 0.0 18.10 0 0.6790 5.304 89.1 1.6475 24 666 20.2 127.36 26.64 10.4
73.53410 0.0 18.10 0 0.6790 5.957 100.0 1.8026 24 666 20.2 16.45 20.62 8.8
11.81230 0.0 18.10 0 0.7180 6.824 76.5 1.7940 24 666 20.2 48.45 22.74 8.4
11.08740 0.0 18.10 0 0.7180 6.411 100.0 1.8589 24 666 20.2 318.75 15.02 16.7
7.02259 0.0 18.10 0 0.7180 6.006 95.3 1.8746 24 666 20.2 319.98 15.70 14.2
12.04820 0.0 18.10 0 0.6140 5.648 87.6 1.9512 24 666 20.2 291.55 14.10 20.8
7.05042 0.0 18.10 0 0.6140 6.103 85.1 2.0218 24 666 20.2 2.52 23.29 13.4
8.79212 0.0 18.10 0 0.5840 5.565 70.6 2.0635 24 666 20.2 3.65 17.16 11.7
15.86030 0.0 18.10 0 0.6790 5.896 95.4 1.9096 24 666 20.2 7.68 24.39 8.3
12.24720 0.0 18.10 0 0.5840 5.837 59.7 1.9976 24 666 20.2 24.65 15.69 10.2
37.66190 0.0 18.10 0 0.6790 6.202 78.7 1.8629 24 666 20.2 18.82 14.52 10.9
7.36711 0.0 18.10 0 0.6790 6.193 78.1 1.9356 24 666 20.2 96.73 21.52 11.0
9.33889 0.0 18.10 0 0.6790 6.380 95.6 1.9682 24 666 20.2 60.72 24.08 9.5
8.49213 0.0 18.10 0 0.5840 6.348 86.1 2.0527 24 666 20.2 83.45 17.64 14.5
10.06230 0.0 18.10 0 0.5840 6.833 94.3 2.0882 24 666 20.2 81.33 19.69 14.1
6.44405 0.0 18.10 0 0.5840 6.425 74.8 2.2004 24 666 20.2 97.95 12.03 16.1
5.58107 0.0 18.10 0 0.7130 6.436 87.9 2.3158 24 666 20.2 100.19 16.22 14.3
13.91340 0.0 18.10 0 0.7130 6.208 95.0 2.2222 24 666 20.2 100.63 15.17 11.7
11.16040 0.0 18.10 0 0.7400 6.629 94.6 2.1247 24 666 20.2 109.85 23.27 13.4
14.42080 0.0 18.10 0 0.7400 6.461 93.3 2.0026 24 666 20.2 27.49 18.05 9.6
15.17720 0.0 18.10 0 0.7400 6.152 100.0 1.9142 24 666 20.2 9.32 26.45 8.7
13.67810 0.0 18.10 0 0.7400 5.935 87.9 1.8206 24 666 20.2 68.95 34.02 8.4
9.39063 0.0 18.10 0 0.7400 5.627 93.9 1.8172 24 666 20.2 396.90 22.88 12.8
22.05110 0.0 18.10 0 0.7400 5.818 92.4 1.8662 24 666 20.2 391.45 22.11 10.5
9.72418 0.0 18.10 0 0.7400 6.406 97.2 2.0651 24 666 20.2 385.96 19.52 17.1
5.66637 0.0 18.10 0 0.7400 6.219 100.0 2.0048 24 666 20.2 395.69 16.59 18.4
9.96654 0.0 18.10 0 0.7400 6.485 100.0 1.9784 24 666 20.2 386.73 18.85 15.4
12.80230 0.0 18.10 0 0.7400 5.854 96.6 1.8956 24 666 20.2 240.52 23.79 10.8
10.67180 0.0 18.10 0 0.7400 6.459 94.8 1.9879 24 666 20.2 43.06 23.98 11.8
6.28807 0.0 18.10 0 0.7400 6.341 96.4 2.0720 24 666 20.2 318.01 17.79 14.9
9.92485 0.0 18.10 0 0.7400 6.251 96.6 2.1980 24 666 20.2 388.52 16.44 12.6
9.32909 0.0 18.10 0 0.7130 6.185 98.7 2.2616 24 666 20.2 396.90 18.13 14.1
7.52601 0.0 18.10 0 0.7130 6.417 98.3 2.1850 24 666 20.2 304.21 19.31 13.0
6.71772 0.0 18.10 0 0.7130 6.749 92.6 2.3236 24 666 20.2 0.32 17.44 13.4
5.44114 0.0 18.10 0 0.7130 6.655 98.2 2.3552 24 666 20.2 355.29 17.73 15.2
5.09017 0.0 18.10 0 0.7130 6.297 91.8 2.3682 24 666 20.2 385.09 17.27 16.1
8.24809 0.0 18.10 0 0.7130 7.393 99.3 2.4527 24 666 20.2 375.87 16.74 17.8
9.51363 0.0 18.10 0 0.7130 6.728 94.1 2.4961 24 666 20.2 6.68 18.71 14.9
4.75237 0.0 18.10 0 0.7130 6.525 86.5 2.4358 24 666 20.2 50.92 18.13 14.1
4.66883 0.0 18.10 0 0.7130 5.976 87.9 2.5806 24 666 20.2 10.48 19.01 12.7
8.20058 0.0 18.10 0 0.7130 5.936 80.3 2.7792 24 666 20.2 3.50 16.94 13.5
7.75223 0.0 18.10 0 0.7130 6.301 83.7 2.7831 24 666 20.2 272.21 16.23 14.9
6.80117 0.0 18.10 0 0.7130 6.081 84.4 2.7175 24 666 20.2 396.90 14.70 20.0
4.81213 0.0 18.10 0 0.7130 6.701 90.0 2.5975 24 666 20.2 255.23 16.42 16.4
3.69311 0.0 18.10 0 0.7130 6.376 88.4 2.5671 24 666 20.2 391.43 14.65 17.7
6.65492 0.0 18.10 0 0.7130 6.317 83.0 2.7344 24 666 20.2 396.90 13.99 19.5
5.82115 0.0 18.10 0 0.7130 6.513 89.9 2.8016 24 666 20.2 393.82 10.29 20.2
7.83932 0.0 18.10 0 0.6550 6.209 65.4 2.9634 24 666 20.2 396.90 13.22 21.4
3.16360 0.0 18.10 0 0.6550 5.759 48.2 3.0665 24 666 20.2 334.40 14.13 19.9
3.77498 0.0 18.10 0 0.6550 5.952 84.7 2.8715 24 666 20.2 22.01 17.15 19.0
4.42228 0.0 18.10 0 0.5840 6.003 94.5 2.5403 24 666 20.2 331.29 21.32 19.1
15.57570 0.0 18.10 0 0.5800 5.926 71.0 2.9084 24 666 20.2 368.74 18.13 19.1
13.07510 0.0 18.10 0 0.5800 5.713 56.7 2.8237 24 666 20.2 396.90 14.76 20.1
4.34879 0.0 18.10 0 0.5800 6.167 84.0 3.0334 24 666 20.2 396.90 16.29 19.9
4.03841 0.0 18.10 0 0.5320 6.229 90.7 3.0993 24 666 20.2 395.33 12.87 19.6
3.56868 0.0 18.10 0 0.5800 6.437 75.0 2.8965 24 666 20.2 393.37 14.36 23.2
4.64689 0.0 18.10 0 0.6140 6.980 67.6 2.5329 24 666 20.2 374.68 11.66 29.8
8.05579 0.0 18.10 0 0.5840 5.427 95.4 2.4298 24 666 20.2 352.58 18.14 13.8
6.39312 0.0 18.10 0 0.5840 6.162 97.4 2.2060 24 666 20.2 302.76 24.10 13.3
4.87141 0.0 18.10 0 0.6140 6.484 93.6 2.3053 24 666 20.2 396.21 18.68 16.7
15.02340 0.0 18.10 0 0.6140 5.304 97.3 2.1007 24 666 20.2 349.48 24.91 12.0
10.23300 0.0 18.10 0 0.6140 6.185 96.7 2.1705 24 666 20.2 379.70 18.03 14.6
14.33370 0.0 18.10 0 0.6140 6.229 88.0 1.9512 24 666 20.2 383.32 13.11 21.4
5.82401 0.0 18.10 0 0.5320 6.242 64.7 3.4242 24 666 20.2 396.90 10.74 23.0
5.70818 0.0 18.10 0 0.5320 6.750 74.9 3.3317 24 666 20.2 393.07 7.74 23.7
5.73116 0.0 18.10 0 0.5320 7.061 77.0 3.4106 24 666 20.2 395.28 7.01 25.0
2.81838 0.0 18.10 0 0.5320 5.762 40.3 4.0983 24 666 20.2 392.92 10.42 21.8
2.37857 0.0 18.10 0 0.5830 5.871 41.9 3.7240 24 666 20.2 370.73 13.34 20.6
3.67367 0.0 18.10 0 0.5830 6.312 51.9 3.9917 24 666 20.2 388.62 10.58 21.2
5.69175 0.0 18.10 0 0.5830 6.114 79.8 3.5459 24 666 20.2 392.68 14.98 19.1
4.83567 0.0 18.10 0 0.5830 5.905 53.2 3.1523 24 666 20.2 388.22 11.45 20.6
0.15086 0.0 27.74 0 0.6090 5.454 92.7 1.8209 4 711 20.1 395.09 18.06 15.2
0.18337 0.0 27.74 0 0.6090 5.414 98.3 1.7554 4 711 20.1 344.05 23.97 7.0
0.20746 0.0 27.74 0 0.6090 5.093 98.0 1.8226 4 711 20.1 318.43 29.68 8.1
0.10574 0.0 27.74 0 0.6090 5.983 98.8 1.8681 4 711 20.1 390.11 18.07 13.6
0.11132 0.0 27.74 0 0.6090 5.983 83.5 2.1099 4 711 20.1 396.90 13.35 20.1
0.17331 0.0 9.69 0 0.5850 5.707 54.0 2.3817 6 391 19.2 396.90 12.01 21.8
0.27957 0.0 9.69 0 0.5850 5.926 42.6 2.3817 6 391 19.2 396.90 13.59 24.5
0.17899 0.0 9.69 0 0.5850 5.670 28.8 2.7986 6 391 19.2 393.29 17.60 23.1
0.28960 0.0 9.69 0 0.5850 5.390 72.9 2.7986 6 391 19.2 396.90 21.14 19.7
0.26838 0.0 9.69 0 0.5850 5.794 70.6 2.8927 6 391 19.2 396.90 14.10 18.3
0.23912 0.0 9.69 0 0.5850 6.019 65.3 2.4091 6 391 19.2 396.90 12.92 21.2
0.17783 0.0 9.69 0 0.5850 5.569 73.5 2.3999 6 391 19.2 395.77 15.10 17.5
0.22438 0.0 9.69 0 0.5850 6.027 79.7 2.4982 6 391 19.2 396.90 14.33 16.8
0.06263 0.0 11.93 0 0.5730 6.593 69.1 2.4786 1 273 21.0 391.99 9.67 22.4
0.04527 0.0 11.93 0 0.5730 6.120 76.7 2.2875 1 273 21.0 396.90 9.08 20.6
0.06076 0.0 11.93 0 0.5730 6.976 91.0 2.1675 1 273 21.0 396.90 5.64 23.9
0.10959 0.0 11.93 0 0.5730 6.794 89.3 2.3889 1 273 21.0 393.45 6.48 22.0
0.04741 0.0 11.93 0 0.5730 6.030 80.8 2.5050 1 273 21.0 396.90 7.88 11.9

References

James, Gareth, Daniela Witten, Trevor Hastie, and Robert Tibshirani. 2013. An Introduction to Statistical Learning with Applications in r. Springer.
Trevor Hastie, Jerome Friedman, Robert Tibshirani. 2009. The Elements of Statistical Learning. 2nd Ed. Springer.