Rejoinder: On nearly assumption-free tests of nominal confidence interval coverage for causal parameters estimated by machine learning

TL;DR

This paper discusses the robustness of confidence interval coverage tests for causal parameters estimated via machine learning, emphasizing minimal assumptions and addressing critiques in the statistical methodology.

Contribution

It provides a response to discussions on nearly assumption-free tests for causal inference, clarifying and defending the proposed methods.

Findings

Highlights the robustness of the proposed tests under minimal assumptions

Addresses critiques and clarifies methodological points

Reinforces the validity of confidence interval coverage in causal estimation

Abstract

This is the rejoinder to the discussion by Kennedy, Balakrishnan and Wasserman on the paper "On nearly assumption-free tests of nominal confidence interval coverage for causal parameters estimated by machine learning" published in Statistical Science.