Leaning Time-Varying Instruments for Identifying Causal Effects in Time-Series Data

TL;DR

This paper introduces TDCIV, a novel method using LSTM and VAE models to learn time-varying instrumental variables from proxy data, enabling unbiased causal effect estimation in complex time-series with latent confounders.

Contribution

It develops the first approach to learn time-varying conditional instrumental variables without domain knowledge, addressing biases from latent confounders in dynamic settings.

Findings

Theoretical validation of the learned representations for causal inference.

TDCIV effectively disentangles and learns time-varying CIVs from proxy variables.

Enables accurate causal effect estimation in complex, real-world time-series data.

Abstract

Querying causal effects from time-series data is important across various fields, including healthcare, economics, climate science, and epidemiology. However, this task becomes complex in the existence of time-varying latent confounders, which affect both treatment and outcome variables over time and can introduce bias in causal effect estimation. Traditional instrumental variable (IV) methods are limited in addressing such complexities due to the need for predefined IVs or strong assumptions that do not hold in dynamic settings. To tackle these issues, we develop a novel Time-varying Conditional Instrumental Variables (CIV) for Debiasing causal effect estimation, referred to as TDCIV. TDCIV leverages Long Short-Term Memory (LSTM) and Variational Autoencoder (VAE) models to disentangle and learn the representations of time-varying CIV and its conditioning set from proxy variables without prior knowledge. Under the assumptions of the Markov property and availability of proxy variables, we theoretically establish the validity of these learned representations for addressing the biases from time-varying latent confounders, thus enabling accurate causal effect estimation. Our proposed TDCIV is the first to effectively learn time-varying CIV and its associated conditioning set without relying on domain-specific knowledge.