Prediction Intervals for Individual Treatment Effects in a Multiple Decision Point Framework using Conformal Inference

TL;DR

This paper introduces a new conformal inference method for constructing prediction intervals for time-varying individual treatment effects across multiple decision points, with theoretical guarantees and real-world applications.

Contribution

It develops a novel conformal inference approach for ITEs that requires weaker assumptions and provides coverage guarantees dependent on data non-exchangeability.

Findings

Method guarantees a lower bound for coverage based on data non-exchangeability.

Simulations show effectiveness in micro-randomized trial settings.

Application to the Intern Health Study demonstrates practical utility.

Abstract

Accurately quantifying uncertainty of individual treatment effects (ITEs) across multiple decision points is crucial for personalized decision-making in fields such as healthcare, finance, education, and online marketplaces. Previous work has focused on predicting non-causal longitudinal estimands or constructing prediction bands for ITEs using cross-sectional data based on exchangeability assumptions. We propose a novel method for constructing prediction intervals using conformal inference techniques for time-varying ITEs with weaker assumptions than prior literature. We guarantee a lower bound for coverage, which is dependent on the degree of non-exchangeability in the data. Although our method is broadly applicable across decision-making contexts, we support our theoretical claims with simulations emulating micro-randomized trials (MRTs) -- a sequential experimental design for mobile health (mHealth) studies. We demonstrate the practical utility of our method by applying it to a real-world MRT - the Intern Health Study (IHS).