Estimating Individual Dose-Response Curves under Unobserved Confounders from Observational Data

TL;DR

ContiVAE is a new framework that estimates individual dose-response curves under unobserved confounders using observational data, advancing causal inference for continuous treatments.

Contribution

It introduces ContiVAE, a variational auto-encoder with a Tilted Gaussian prior, capable of modeling hidden confounders and predicting individual responses to continuous treatments.

Findings

ContiVAE outperforms existing methods by up to 62% on semi-synthetic datasets.

It effectively captures heterogeneity among individuals.

Demonstrates practical utility on real-world data.

Abstract

Estimating an individual's potential response to continuously varied treatments is crucial for addressing causal questions across diverse domains, from healthcare to social sciences. However, existing methods are limited either to estimating causal effects of binary treatments, or scenarios where all confounding variables are measurable. In this work, we present ContiVAE, a novel framework for estimating causal effects of continuous treatments, measured by individual dose-response curves, considering the presence of unobserved confounders using observational data. Leveraging a variational auto-encoder with a Tilted Gaussian prior distribution, ContiVAE models the hidden confounders as latent variables, and is able to predict the potential outcome of any treatment level for each individual while effectively capture the heterogeneity among individuals. Experiments on semi-synthetic datasets show that ContiVAE outperforms existing methods by up to 62%, demonstrating its robustness and flexibility. Application on a real-world dataset illustrates its practical utility.