Causal Dynamic Variational Autoencoder for Counterfactual Regression in Longitudinal Data

TL;DR

This paper introduces CDVAE, a novel method for estimating treatment effects over time by modeling unobserved confounders, demonstrating improved accuracy on synthetic and real data.

Contribution

The paper proposes a causal dynamic variational autoencoder that captures unobserved confounders for better treatment effect estimation in longitudinal data.

Findings

CDVAE outperforms existing methods on synthetic datasets.

Augmenting models with CDVAE's latent variables improves CATE estimates.

Approaches approach oracle performance without access to true confounders.

Abstract

Accurately estimating treatment effects over time is crucial in fields such as precision medicine, epidemiology, economics, and marketing. Many current methods for estimating treatment effects over time assume that all confounders are observed or attempt to infer unobserved ones. In contrast, our approach focuses on unobserved adjustment variables, which specifically have a causal effect on the outcome sequence. Under the assumption of unconfoundedness, we address estimating Conditional Average Treatment Effects (CATEs) while accounting for unobserved heterogeneity in response to treatment due to these unobserved adjustment variables. Our proposed Causal Dynamic Variational Autoencoder (CDVAE) is grounded in theoretical guarantees concerning the validity of latent adjustment variables and generalization bounds on CATE estimation error. Extensive evaluations on synthetic and real-world datasets show that CDVAE outperforms existing baselines. Moreover, we demonstrate that state-of-the-art models significantly improve their CATE estimates when augmented with the latent substitutes learned by CDVAE, approaching oracle-level performance without direct access to the true adjustment variables.