Reference implementation of the medical deconfounder

This repository contains code for Zhang et al., 2019.

It implements the two-cause simulation (Section 3.2) and the multi-cause simulation (Section 3.3).

System requirements

1. Python 2.7
2. Edward 1.3.5
3. tensorflow 1.5.0
4. R 3.6.0
5. rstan 2.18.2
6. rstanarm 2.18.2

References

L. Zhang, Y. Wang, A. Ostropolets, J.J. Mulgrave, D.M. Blei, and G. Hripcsak. The Medical Deconfounder: Assessing Treatment Effects with Electronic Health Records. In Proceedings of the 4th Machine Learning for Healthcare Conference, volume 106 of Proceedings of Machine Learning Research, pages 490-512, Ann Arbor, Michigan, 2019. PDF