Causal Effect Estimation with Learned Instrument Representations

TL;DR

This paper introduces ZNet, a representation learning model that constructs instrumental variables from observed data, enabling causal effect estimation without explicit instruments in observational studies.

Contribution

ZNet is a novel architecture that decomposes features into confounding and instrumental parts, facilitating IV-based causal inference without predefined instruments.

Findings

ZNet can recover true instruments when they exist in data.

It can create latent instruments in the embedding space without explicit IVs.

Experiments show ZNet's effectiveness in various causal inference scenarios.

Abstract

Instrumental variable (IV) methods mitigate bias from unobserved confounding in observational causal inference but rely on the availability of a valid instrument, which can often be difficult or infeasible to identify in practice. In this paper, we propose a representation learning approach that constructs instrumental representations from observed covariates, which enable IV-based estimation even in the absence of an explicit instrument. Our model (ZNet) achieves this through an architecture that mirrors the structural causal model of IVs; it decomposes the ambient feature space into confounding and instrumental components, and is trained by enforcing empirical moment conditions corresponding to the defining properties of valid instruments (i.e., relevance, exclusion restriction, and instrumental unconfoundedness). Importantly, ZNet is compatible with a wide range of downstream two-stage IV estimators of causal effects. Our experiments demonstrate that ZNet can (i) recover ground-truth instruments when they already exist in the ambient feature space and (ii) construct latent instruments in the embedding space when no explicit IVs are available. Our work suggests when ZNet can be used as a module for causal inference in general observational settings.