Conformal Meta-learners for Predictive Inference of Individual Treatment Effects

TL;DR

This paper introduces conformal meta-learners for predictive inference of individual treatment effects, providing valid predictive intervals that are robust and applicable to a broad class of CATE meta-learners, with strong theoretical guarantees and practical performance.

Contribution

It develops a general conformal inference framework for ITEs based on CATE meta-learners, ensuring valid predictive intervals with minimal assumptions.

Findings

Conformal meta-learners produce valid predictive intervals for ITEs.

The method is applicable to a broad class of CATE meta-learners, including doubly-robust learners.

Numerical experiments demonstrate competitive efficiency and validity of the intervals.

Abstract

We investigate the problem of machine learning-based (ML) predictive inference on individual treatment effects (ITEs). Previous work has focused primarily on developing ML-based meta-learners that can provide point estimates of the conditional average treatment effect (CATE); these are model-agnostic approaches for combining intermediate nuisance estimates to produce estimates of CATE. In this paper, we develop conformal meta-learners, a general framework for issuing predictive intervals for ITEs by applying the standard conformal prediction (CP) procedure on top of CATE meta-learners. We focus on a broad class of meta-learners based on two-stage pseudo-outcome regression and develop a stochastic ordering framework to study their validity. We show that inference with conformal meta-learners is marginally valid if their (pseudo outcome) conformity scores stochastically dominate oracle conformity scores evaluated on the unobserved ITEs. Additionally, we prove that commonly used CATE meta-learners, such as the doubly-robust learner, satisfy a model- and distribution-free stochastic (or convex) dominance condition, making their conformal inferences valid for practically-relevant levels of target coverage. Whereas existing procedures conduct inference on nuisance parameters (i.e., potential outcomes) via weighted CP, conformal meta-learners enable direct inference on the target parameter (ITE). Numerical experiments show that conformal meta-learners provide valid intervals with competitive efficiency while retaining the favorable point estimation properties of CATE meta-learners.