Bounds on Causal Effects and Application to High Dimensional Data
This work addresses causal inference challenges in high-dimensional data for researchers and practitioners, but it appears incremental as it builds on existing criteria with a new optimization-based approach.
The paper tackles the problem of estimating causal effects when adjustment variables are partially observed by deriving bounds through non-linear optimization, and proposes a dimensionality reduction framework that trades bias for estimation power, with performance demonstrated in simulations.
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.