因果推断学习笔记:资料收集
被几个causal inference的tutorial拉入了坑,觉得还蛮有意思的。 写一篇笔记收藏一些自己找到的的资料吧,持续更新中。
科普文章
- 统计之都上有一个因果推断系列,写得很好
教材
- Causality: Models, Reasoning and Inference. Judea Pearl. 偏重 causal diagram 一些。
- Causal Inference for Statistics, Social, and Biomedical. Sciences: An Introduction. Guido W. Imbens and. Donald B. Rubin. 个人感觉比较实用。
- Mostly harmless econometrics: An empiricist’s companion. Angrist, J. D., & Pischke, J. S.. 实用。作者对银河系漫游指南真的是真爱。
- Causal Inference: The Mixtape. Scott Cunningham. 可以在教授主页上下载到 PDF。内容 self-contained,很全。
公开课
带视频的公开课
- Four Lectures on Causality (Prof. Jonas Peters, University of Copenhagen @MIT)
- 主页: https://stat.mit.edu/news/four-lectures-causality/
- 偏理论一些,和Pearl一个流派
- A Crash Course in Causality: Inferring Causal Effects from Observational Data (Prof. Jason A. Roy, Ph.D., UPenn)
- 主页: https://www.coursera.org/learn/crash-course-in-causality
- 入门课程,通俗易懂,带有一部分R的tutorial和tests可以练手
只有PPT的公开课
- Applied Causality (Spring 2017, Columbia University, David M. Blei)
Tutorial
- Tutorial on Causal Inference and Counterfactual Reasoning (KDD’18)
- Presenter: Amit Sharma (@amt_shrma), Emre Kiciman (@emrek)
- 主页:https://causalinference.gitlab.io/kdd-tutorial/
- Tutorial on Counterfactual Inference (NIPS’18)
- Presenter: Susan Athey (Stanford)
- 视频:https://www.facebook.com/nipsfoundation/videos/1291139774361116/UzpfSTEwMDAwMTk2OTU5ODE5NToxOTQwNTU1NTUyNjg2NzQ2/
- Slides和其它资料: /https://drive.google.com/drive/folders/1SEEOMluxBcSAb_tsDYgcLFtOQaeWtkLp
- Machine Learning and Econometrics (2018 AEA Continuing Education Webcasts)
- Presenter: Susan Athey, Guido Imbens
- 视频:https://www.aeaweb.org/conference/cont-ed/2018-webcasts
- Slides和其它资料: /https://drive.google.com/drive/folders/1SEEOMluxBcSAb_tsDYgcLFtOQaeWtkLp
TODO
- Statistical and causal inference in social networks (WWW2015 School)
工业界应用收集
收集一些作者来自工业界的PPT。
- _Causality without headaches_ (Benoît Rostykus, Senior Machine Learning Researcher at Netflix) https://www.slideshare.net/BenoitRostykus/causality-without-headaches
R Packages
- MatchIt:各种matching方法。
- WeightIt:封装了各种weighting方法。
- grf (generalized random forests)
- bartCause
其它
- Github rguo12/awesome-causality-algorithms:paper和code列表
- Github rguo12/awesome-causality-data:数据集列表
- Causality paper @Totte Harinen:一位 Uber data scientist 的博客