Forecasting Causal Effects of Future Interventions: Confounding and Transportability Issues

2024-09-19

AI Summary

This study addresses the challenge of cross-temporal causal extrapolation: reliably predicting the causal effects of future interventions (e.g., public health policies) in dynamically evolving populations using historical observational data—particularly under time-varying confounding and effect modification. To this end, we propose the first nonparametric g-computation identification formula specifically designed for cross-temporal prediction, rigorously specifying identifiability conditions, the target estimand, and structural assumptions. Our approach integrates the potential outcomes framework, temporal causal modeling, and modern causal inference theory to establish a sound theoretical foundation. We validate the feasibility and practical utility of the method through an empirical analysis of the impact of COVID-19 policies on mortality. The results demonstrate its applicability in dynamic, real-world settings and provide a generalizable theoretical framework and computationally tractable pathway for causal forecasting under temporal heterogeneity.

Technology Category

• Reasoning under Uncertainty: Causality
• Machine Learning: Causal Learning
• Knowledge Representation and Reasoning: Action, Change, and Causality

Application Category

• Graph Algorithms and Modeling for the Web: Algorithms and analysis for heterogeneous, signed, attributed, multi-relational, temporal, higher-order, and annotated Web-related graphs
• User Modeling, Personalization and Recommendation: Attacks and countermeasures in recommendation systems
• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking

Abstract

Recent developments in causal inference allow us to transport a causal effect of a time-fixed treatment from a randomized trial to a target population across space but within the same time frame. In contrast to transportability across space, transporting causal effects across time or forecasting causal effects of future interventions is more challenging due to time-varying confounders and time-varying effect modifiers. In this article, we seek to formally clarify the causal estimands for forecasting causal effects over time and the structural assumptions required to identify these estimands. Specifically, we develop a set of novel nonparametric identification formulas--g-computation formulas--for these causal estimands, and lay out the conditions required to accurately forecast causal effects from a past observed sample to a future population in a future time window. Our overaching objective is to leverage the modern causal inference theory to provide a theoretical framework for investigating whether the effects seen in a past sample would carry over to a new future population. Throughout the article, a working example addressing the effect of public policies or social events on COVID-related deaths is considered to contextualize the developments of analytical results.

Problem

• Forecasting causal effects of future interventions over time
• Addressing time-varying confounders and effect modifiers
• Developing identification formulas for future causal estimands

Innovation

• Nonparametric g-computation formulas for causal estimands
• Addressing time-varying confounders and modifiers
• Transporting causal effects across time frames