Saved in:
Bibliographic Details
Main Authors: Schur, Felix, Pfister, Niklas, Ding, Peng, Mukherjee, Sach, Peters, Jonas
Format: Preprint
Published: 2026
Subjects:
Online Access:https://arxiv.org/abs/2601.15254
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866912838523551744
author Schur, Felix
Pfister, Niklas
Ding, Peng
Mukherjee, Sach
Peters, Jonas
author_facet Schur, Felix
Pfister, Niklas
Ding, Peng
Mukherjee, Sach
Peters, Jonas
contents We study the problem of estimating causal effects under hidden confounding in the following unpaired data setting: we observe some covariates $X$ and an outcome $Y$ under different experimental conditions (environments) but do not observe them jointly; we either observe $X$ or $Y$. Under appropriate regularity conditions, the problem can be cast as an instrumental variable (IV) regression with the environment acting as a (possibly high-dimensional) instrument. When there are many environments but only a few observations per environment, standard two-sample IV estimators fail to be consistent. We propose a GMM-type estimator based on cross-fold sample splitting of the instrument-covariate sample and prove that it is consistent as the number of environments grows but the sample size per environment remains constant. We further extend the method to sparse causal effects via $\ell_1$-regularized estimation and post-selection refitting.
format Preprint
id arxiv_https___arxiv_org_abs_2601_15254
institution arXiv
publishDate 2026
record_format arxiv
spellingShingle Many Experiments, Few Repetitions, Unpaired Data, and Sparse Effects: Is Causal Inference Possible?
Schur, Felix
Pfister, Niklas
Ding, Peng
Mukherjee, Sach
Peters, Jonas
Machine Learning
Artificial Intelligence
We study the problem of estimating causal effects under hidden confounding in the following unpaired data setting: we observe some covariates $X$ and an outcome $Y$ under different experimental conditions (environments) but do not observe them jointly; we either observe $X$ or $Y$. Under appropriate regularity conditions, the problem can be cast as an instrumental variable (IV) regression with the environment acting as a (possibly high-dimensional) instrument. When there are many environments but only a few observations per environment, standard two-sample IV estimators fail to be consistent. We propose a GMM-type estimator based on cross-fold sample splitting of the instrument-covariate sample and prove that it is consistent as the number of environments grows but the sample size per environment remains constant. We further extend the method to sparse causal effects via $\ell_1$-regularized estimation and post-selection refitting.
title Many Experiments, Few Repetitions, Unpaired Data, and Sparse Effects: Is Causal Inference Possible?
topic Machine Learning
Artificial Intelligence
url https://arxiv.org/abs/2601.15254