CAETC: Causal Autoencoding and Treatment Conditioning for Counterfactual Estimation over Time

TL;DR

CAETC is a novel method that uses autoencoding and treatment conditioning to improve counterfactual estimation over time, effectively addressing time-dependent confounding bias in observational data.

Contribution

It introduces a treatment-invariant autoencoding framework that enhances counterfactual predictions in time-series data, applicable across various sequence models.

Findings

Significant improvement over existing methods in synthetic data

Effective in semi-synthetic and real-world datasets

Robust to different sequence modeling architectures

Abstract

Counterfactual estimation over time is important in various applications, such as personalized medicine. However, time-dependent confounding bias in observational data still poses a significant challenge in achieving accurate and efficient estimation. We introduce causal autoencoding and treatment conditioning (CAETC), a novel method for this problem. Built on adversarial representation learning, our method leverages an autoencoding architecture to learn a partially invertible and treatment-invariant representation, where the outcome prediction task is cast as applying a treatment-specific conditioning on the representation. Our design is independent of the underlying sequence model and can be applied to existing architectures such as long short-term memories (LSTMs) or temporal convolution networks (TCNs). We conduct extensive experiments on synthetic, semi-synthetic, and real-world data to demonstrate that CAETC yields significant improvement in counterfactual estimation over existing methods.