Bounds on Causal Effects and Application to High Dimensional Data

Abstract

This paper addresses the problem of estimating causal effects when adjustment variables in the back-door or front-door criterion are partially observed. For such scenarios, we derive bounds on the causal effects by solving two non-linear optimization problems, and demonstrate that the bounds are sufficient. Using this optimization method, we propose a framework for dimensionality reduction that allows one to trade bias for estimation power, and demonstrate its performance using simulation studies.

Cite

Text

Li and Pearl. "Bounds on Causal Effects and Application to High Dimensional Data." AAAI Conference on Artificial Intelligence, 2022. doi:10.1609/AAAI.V36I5.20520

Markdown

[Li and Pearl. "Bounds on Causal Effects and Application to High Dimensional Data." AAAI Conference on Artificial Intelligence, 2022.](https://mlanthology.org/aaai/2022/li2022aaai-bounds/) doi:10.1609/AAAI.V36I5.20520

BibTeX

@inproceedings{li2022aaai-bounds,
  title     = {{Bounds on Causal Effects and Application to High Dimensional Data}},
  author    = {Li, Ang and Pearl, Judea},
  booktitle = {AAAI Conference on Artificial Intelligence},
  year      = {2022},
  pages     = {5773-5780},
  doi       = {10.1609/AAAI.V36I5.20520},
  url       = {https://mlanthology.org/aaai/2022/li2022aaai-bounds/}
}