Orthogonal prediction of counterfactual outcomes

TL;DR

This paper introduces orthogonal meta-learners for counterfactual outcome prediction that respect outcome constraints, improving performance especially for dichotomous outcomes, validated through simulations and critical care data analysis.

Contribution

It develops orthogonal meta-learners that incorporate outcome space constraints, enhancing counterfactual prediction accuracy over existing methods.

Findings

Outperforms existing meta-learners in simulations

Effective for dichotomous outcomes with constrained spaces

Validated on critical care dataset

Abstract

Orthogonal meta-learners, such as DR-learner, R-learner and IF-learner, are increasingly used to estimate conditional average treatment effects. They improve convergence rates relative to na\"{\i}ve meta-learners (e.g., T-, S- and X-learner) through de-biasing procedures that involve applying standard learners to specifically transformed outcome data. This leads them to disregard the possibly constrained outcome space, which can be particularly problematic for dichotomous outcomes: these typically get transformed to values that are no longer constrained to the unit interval, making it difficult for standard learners to guarantee predictions within the unit interval. To address this, we construct orthogonal meta-learners for the prediction of counterfactual outcomes which respect the outcome space. As such, the obtained i-learner or imputation-learner is more generally expected to outperform existing learners, even when the outcome is unconstrained, as we confirm empirically in simulation studies and an analysis of critical care data. Our development also sheds broader light onto the construction of orthogonal learners for other estimands.