Cross-Validated Causal Inference: a Modern Method to Combine Experimental and Observational Data

Fuente: arXiv
Enregistré dans:
Détails bibliographiques
Auteurs principaux: Yang, Xuelin, Lin, Licong, Athey, Susan, Jordan, Michael I., Imbens, Guido W.
Format: Preprint
Publié: 2025
Sujets:
Accès en ligne:
Tags: Ajouter un tag
Pas de tags, Soyez le premier à ajouter un tag!
_version_ 1866918181837209600
author Yang, Xuelin
Lin, Licong
Athey, Susan
Jordan, Michael I.
Imbens, Guido W.
author_facet Yang, Xuelin
Lin, Licong
Athey, Susan
Jordan, Michael I.
Imbens, Guido W.
contents We develop new methods to integrate experimental and observational data in causal inference. While randomized controlled trials offer strong internal validity, they are often costly and therefore limited in sample size. Observational data, though cheaper and often with larger sample sizes, are prone to biases due to unmeasured confounders. To harness their complementary strengths, we propose a systematic framework that formulates causal estimation as an empirical risk minimization (ERM) problem. A full model containing the causal parameter is obtained by minimizing a weighted combination of experimental and observational losses--capturing the causal parameter's validity and the full model's fit, respectively. The weight is chosen through cross-validation on the causal parameter across experimental folds. Our experiments on real and synthetic data show the efficacy and reliability of our method. We also provide theoretical non-asymptotic error bounds.
format Preprint
id arxiv_https___arxiv_org_abs_2511_00727
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Cross-Validated Causal Inference: a Modern Method to Combine Experimental and Observational Data
Yang, Xuelin
Lin, Licong
Athey, Susan
Jordan, Michael I.
Imbens, Guido W.
Econometrics
Methodology
Machine Learning
We develop new methods to integrate experimental and observational data in causal inference. While randomized controlled trials offer strong internal validity, they are often costly and therefore limited in sample size. Observational data, though cheaper and often with larger sample sizes, are prone to biases due to unmeasured confounders. To harness their complementary strengths, we propose a systematic framework that formulates causal estimation as an empirical risk minimization (ERM) problem. A full model containing the causal parameter is obtained by minimizing a weighted combination of experimental and observational losses--capturing the causal parameter's validity and the full model's fit, respectively. The weight is chosen through cross-validation on the causal parameter across experimental folds. Our experiments on real and synthetic data show the efficacy and reliability of our method. We also provide theoretical non-asymptotic error bounds.
title Cross-Validated Causal Inference: a Modern Method to Combine Experimental and Observational Data
topic Econometrics
Methodology
Machine Learning
url https://arxiv.org/abs/2511.00727