Causal Inference from Small High-dimensional Datasets

TL;DR

This paper introduces Causal-Batle, a transfer learning-based method designed to improve treatment effect estimation in small, high-dimensional datasets, addressing the limitations of neural network estimators in such scenarios.

Contribution

It presents a novel transfer learning approach for causal inference tailored to small high-dimensional datasets, enhancing stability and accuracy of treatment effect estimates.

Findings

Transfer learning improves neural network stability in small datasets.

Causal-Batle outperforms traditional methods in high-dimensional settings.

Enhanced treatment effect estimation accuracy demonstrated in experiments.

Abstract

Many methods have been proposed to estimate treatment effects with observational data. Often, the choice of the method considers the application's characteristics, such as type of treatment and outcome, confounding effect, and the complexity of the data. These methods implicitly assume that the sample size is large enough to train such models, especially the neural network-based estimators. What if this is not the case? In this work, we propose Causal-Batle, a methodology to estimate treatment effects in small high-dimensional datasets in the presence of another high-dimensional dataset in the same feature space. We adopt an approach that brings transfer learning techniques into causal inference. Our experiments show that such an approach helps to bring stability to neural network-based methods and improve the treatment effect estimates in small high-dimensional datasets.