Estimate Time-Varying Exposure Effects via Ensemble Learning-based Marginal Structural Model with Application to Adolescent Cognitive Development Study

2025-10-17

AI Summary

Estimating causal effects of time-varying exposures (e.g., adolescent sleep deprivation) in longitudinal studies is challenged by model misspecification and bias due to high-dimensional, nonlinear time-varying confounders. To address this, we propose Marginal Structural Models via Ensemble learning (MASE), a novel framework that jointly models propensity scores and conditional mean sequences using an ensemble of machine learning algorithms, enabling robust adjustment for hundreds of dynamic confounders. MASE integrates inverse-probability weighting, G-computation, and targeted maximum likelihood estimation. Simulation studies demonstrate that MASE substantially reduces estimation bias, improves confidence interval coverage, and enhances statistical power compared to conventional methods. Empirical analysis reveals a significant cumulative negative effect of sleep deprivation on adolescent cognitive development. The proposed framework provides a scalable, model-robust tool for complex longitudinal causal inference.

Technology Category

• Machine Learning: Causal Learning
• Reasoning under Uncertainty: Causality
• Multiagent Systems: Mechanism Design

Application Category

• User Modeling, Personalization and Recommendation: Studies of user behavior, including longitudinal effects of personalized systems
• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking
• Graph Algorithms and Modeling for the Web: Efficient manipulation of static and dynamic Web-related graphs

Abstract

Evaluating the effects of time-varying exposures is essential for longitudinal studies. The effect estimation becomes increasingly challenging when dealing with hundreds of time-dependent confounders. We propose a Marginal Structure Ensemble Learning Model (MASE) to provide a marginal structure model (MSM)-based robust estimator under the longitudinal setting. The proposed model integrates multiple machine learning algorithms to model propensity scores and a sequence of conditional outcome means such that it becomes less sensitive to model mis-specification due to any single algorithm and allows many confounders with potential non-linear confounding effects to reduce the risk of inconsistent estimation. Extensive simulation analysis demonstrates the superiority of MASE over benchmark methods (e.g., MSM, G-computation, Targeted maximum likelihood), yielding smaller estimation bias and improved inference accuracy. We apply MASE to the adolescent cognitive development study to investigate the time-varying effects of sleep insufficiency on cognitive performance. The results reveal an aggregated negative impact of insufficient sleep on cognitive development among youth.

Problem

• Estimating time-varying exposure effects with many confounders
• Reducing model mis-specification risk in longitudinal studies
• Evaluating sleep insufficiency effects on adolescent cognition

Innovation

• Ensemble learning integrates multiple machine learning algorithms
• Models propensity scores and conditional outcome means
• Reduces bias in time-varying exposure effect estimation