Flexible Nonparametric Inference for Causal Effects under the Front-Door Model

Name: Flexible Nonparametric Inference for Causal Effects under the Front-Door Model
Uploaded: 2026-10-06T02:53:51.690835+00:00
Duration: 558 s
Description: The research develops flexible machine learning estimators to estimate causal treatment effects (ATE/ATT) using the front-door criterion. It introduces novel minimum loss-based estimators, like TMLEs, that are robust against model misspecification and provide strong theoretical guarantees for observational studies.

The research develops flexible machine learning estimators to estimate causal treatment effects (ATE/ATT) using the front-door criterion. It introduces novel minimum loss-based estimators, like TMLEs, that are robust against model misspecification and provide strong theoretical guarantees for observational studies.

Open the full episode