Class flipping for uplift modeling and Heterogeneous Treatment Effect estimation on imbalanced RCT data

TL;DR

This paper introduces a class flipping method for uplift modeling and Heterogeneous Treatment Effect estimation in imbalanced RCT data, avoiding distortion of effects and calibration issues, especially useful for class transformation models.

Contribution

It proposes a novel class flipping approach that preserves effect estimates and addresses class imbalance without calibration, applicable to both uplift modeling and standard classification.

Findings

The method maintains correct effect predictions in imbalanced data.

Experimental results confirm theoretical guarantees.

The approach is effective for class transformation models and standard classification.

Abstract

Uplift modeling and Heterogeneous Treatment Effect (HTE) estimation aim at predicting the causal effect of an action, such as a medical treatment or a marketing campaign on a specific individual. In this paper, we focus on data from Randomized Controlled Experiments which guarantee causal interpretation of the outcomes. Class and treatment imbalance are important problems in uplift modeling/HTE, but classical undersampling or oversampling based approaches are hard to apply in this case since they distort the predicted effect. Calibration methods have been proposed in the past, however, they do not guarantee correct predictions. In this work, we propose an approach alternative to undersampling, based on flipping the class value of selected records. We show that the proposed approach does not distort the predicted effect and does not require calibration. The method is especially useful for models based on class variable transformation (modified outcome models). We address those models separately, designing a transformation scheme which guarantees correct predictions and addresses also the problem of treatment imbalance which is especially important for those models. Experiments fully confirm our theoretical results. Additionally, we demonstrate that our method is a viable alternative also for standard classification problems.