Causal Responder Detection

TL;DR

This paper presents CARD, a new method for responder analysis that uses conformal prediction and propensity score adjustment to accurately identify treatment responders while controlling false discoveries, especially in observational studies.

Contribution

CARD introduces a novel combination of conformal prediction and propensity score adjustment for responder detection, improving robustness and accuracy over existing methods.

Findings

Effectively detects responders with high power in simulations.

Controls false discovery rate in finite samples.

Robust in observational study settings.

Abstract

We introduce the causal responders detection (CARD), a novel method for responder analysis that identifies treated subjects who significantly respond to a treatment. Leveraging recent advances in conformal prediction, CARD employs machine learning techniques to accurately identify responders while controlling the false discovery rate in finite sample sizes. Additionally, we incorporate a propensity score adjustment to mitigate bias arising from non-random treatment allocation, enhancing the robustness of our method in observational settings. Simulation studies demonstrate that CARD effectively detects responders with high power in diverse scenarios.