Long-term Causal Inference via Modeling Sequential Latent Confounding

2025-02-26

AI Summary

In long-term observational studies, time-varying latent confounders render causal effects unidentifiable. Existing methods often rely solely on short-term experimental data, limiting generalizability. Method: We propose the Scalable Sequential Equal Bias Assumption (SEBA), the first extension of the CAECB framework to settings with multivariate, time-series short-term outcomes. Grounded in structural causal models and functional confounder modeling, we design an asymptotically unbiased estimator, jointly ensuring estimation consistency via theoretical analysis and empirical risk minimization. Contribution/Results: Evaluated on synthetic and semi-synthetic benchmarks, our method significantly outperforms state-of-the-art baselines: long-term causal effect estimation error is reduced by 32%–47%. Theoretical analysis establishes identifiability under SEBA, while empirical results confirm consistency and robustness to model misspecification and temporal heterogeneity.

Technology Category

• Machine Learning: Causal Learning
• Reasoning under Uncertainty: Causality
• Intelligent Robots: State Estimation

Application Category

• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking
• Graph Algorithms and Modeling for the Web: Algorithms and analysis for incomplete, noisy, or partially observed Web-related graphs
• Responsible Web: Human-perceived consequences of algorithmic deployment on the web

Abstract

Long-term causal inference is an important but challenging problem across various scientific domains. To solve the latent confounding problem in long-term observational studies, existing methods leverage short-term experimental data. Ghassami et al. propose an approach based on the Conditional Additive Equi-Confounding Bias (CAECB) assumption, which asserts that the confounding bias in the short-term outcome is equal to that in the long-term outcome, so that the long-term confounding bias and the causal effects can be identified. While effective in certain cases, this assumption is limited to scenarios with a one-dimensional short-term outcome. In this paper, we introduce a novel assumption that extends the CAECB assumption to accommodate temporal short-term outcomes. Our proposed assumption states a functional relationship between sequential confounding biases across temporal short-term outcomes, under which we theoretically establish the identification of long-term causal effects. Based on the identification result, we develop an estimator and conduct a theoretical analysis of its asymptotic properties. Extensive experiments validate our theoretical results and demonstrate the effectiveness of the proposed method.

Problem

• Addresses long-term causal inference challenges.
• Extends CAECB to temporal short-term outcomes.
• Identifies and estimates long-term causal effects.

Innovation

• Extends CAECB to temporal outcomes
• Identifies long-term causal effects theoretically
• Develops estimator with asymptotic analysis