Doubly Robust Adaptive Conformal Inference for Causal Effects Under Temporal Dependence

2026-06-29

AI Summary

This study addresses the challenge of quantifying uncertainty in causal effects from time-dependent data by proposing a novel approach that integrates doubly robust estimation with adaptive conformal inference. The method introduces doubly robustness into the time series causal inference framework for the first time and constructs predictive intervals for causal effects using adaptive conformal prediction. By leveraging the strengths of both components, the proposed procedure achieves valid coverage guarantees while substantially improving interval sharpness. Consequently, it enables effective and reliable uncertainty quantification for causal effects under temporal dependence.

Abstract

We propose doubly robust adaptive conformal inference (DR-ACI), which constructs prediction intervals for doubly robust pseudo-outcomes under temporal dependence.

Problem

• causal effects
• temporal dependence
• prediction intervals
• conformal inference
• doubly robust

Innovation

• doubly robust
• adaptive conformal inference
• temporal dependence
• causal effects
• prediction intervals