TFP CausalImpact:一个Python包,用于估计设计干预对时间序列的因果效应
【TFP CausalImpact:一个Python包,用于估计设计干预对时间序列的因果效应,例如,广告活动每天生成了多少额外的点击量。在没有随机实验可用的情况下,回答这样的问题可能很困难。这个包使用结构贝叶斯时间序列模型来估计如果没有进行干预,响应度量可能在干预后如何发展】'TFP CausalImpact'
GitHub - google/tfp-causalimpact
This Python package implements an approach to estimating the causal effect of a designed intervention on a time series. For example, how many additional daily clicks were generated by an advertising campaign? Answering a question like this can be difficult when a randomized experiment is not available. The package aims to address this difficulty using a structural Bayesian time-series model to estimate how the response metric might have evolved after the intervention if the intervention had not occurred [1].
As with all approaches to causal inference on non-experimental data, valid conclusions require strong assumptions. The CausalImpact package, in particular, assumes that the outcome time series can be explained in terms of a set of control time series that were themselves not affected by the intervention. Furthermore, the relation between treated series and control series is assumed to be stable during the post-intervention period. Understanding and checking these assumptions for any given application is critical for obtaining valid conclusions.
TFP CausalImpact is a Python + TensorFlow Probability implementation of the CausalImpact R package developed at Google by Kay Brodersen and Alain Hauser. TFP CausalImpact is based on both the original R package and on a Python version https://github.com/dafiti/causalimpact developed at Dafiti by Willian Fuks. TFP CausalImpact was developed at Google by Colin Carroll, David Moore, Jacob Burnim, Kyle Loveless, and Susanna Makela.
This is not an officially supported Google product.
[1] Inferring causal impact using Bayesian structural time-series models. Kay H. Brodersen, Fabian Gallusser, Jim Koehler, Nicolas Remy, Steven L. Scott. Annals of Applied Statistics, vol. 9 (2015), pp. 247-274. https://research.google/pubs/pub41854/