Learning Heterogeneity in Causal Inference Using Sufficient Dimension Reduction

Often the research interest in causal inference is on the regression causal effect, which is the mean difference in the potential outcomes conditional on the covariates. In this paper, we use sufficient dimension reduction to estimate a lower dimensional linear combination of the covariates that is...

Full description

Bibliographic Details
Main Authors:	Luo Wei, Wu Wenbo, Zhu Yeying
Format:	Article
Language:	English
Published:	De Gruyter 2019-04-01
Series:	Journal of Causal Inference
Subjects:	central causal effect subspace conditional causal effect heterogeneity variable selection
Online Access:	https://doi.org/10.1515/jci-2018-0015

id	doaj-538727ae3dfe4c5fade10a46ce3e837c
record_format	Article
spelling	doaj-538727ae3dfe4c5fade10a46ce3e837c2021-09-06T19:40:28ZengDe GruyterJournal of Causal Inference2193-36772193-36852019-04-017168870110.1515/jci-2018-0015Learning Heterogeneity in Causal Inference Using Sufficient Dimension ReductionLuo Wei0Wu Wenbo1Zhu Yeying2Center for Data Science, Zhejiang University, Hangzhou, ChinaDepartment of Management Science and Statistics, The University of Texas at San Antonio, San Antonio, United StatesDepartment of Statistics and Actuarial Science, University of Waterloo, Waterloo, CanadaOften the research interest in causal inference is on the regression causal effect, which is the mean difference in the potential outcomes conditional on the covariates. In this paper, we use sufficient dimension reduction to estimate a lower dimensional linear combination of the covariates that is sufficient to model the regression causal effect. Compared with the existing applications of sufficient dimension reduction in causal inference, our approaches are more efficient in reducing the dimensionality of covariates, and avoid estimating the individual outcome regressions. The proposed approaches can be used in three ways to assist modeling the regression causal effect: to conduct variable selection, to improve the estimation accuracy, and to detect the heterogeneity. Their usefulness are illustrated by both simulation studies and a real data example.https://doi.org/10.1515/jci-2018-0015central causal effect subspaceconditional causal effectheterogeneityvariable selection
collection	DOAJ
language	English
format	Article
sources	DOAJ
author	Luo Wei Wu Wenbo Zhu Yeying
spellingShingle	Luo Wei Wu Wenbo Zhu Yeying Learning Heterogeneity in Causal Inference Using Sufficient Dimension Reduction Journal of Causal Inference central causal effect subspace conditional causal effect heterogeneity variable selection
author_facet	Luo Wei Wu Wenbo Zhu Yeying
author_sort	Luo Wei
title	Learning Heterogeneity in Causal Inference Using Sufficient Dimension Reduction
title_short	Learning Heterogeneity in Causal Inference Using Sufficient Dimension Reduction
title_full	Learning Heterogeneity in Causal Inference Using Sufficient Dimension Reduction
title_fullStr	Learning Heterogeneity in Causal Inference Using Sufficient Dimension Reduction
title_full_unstemmed	Learning Heterogeneity in Causal Inference Using Sufficient Dimension Reduction
title_sort	learning heterogeneity in causal inference using sufficient dimension reduction
publisher	De Gruyter
series	Journal of Causal Inference
issn	2193-3677 2193-3685
publishDate	2019-04-01
description	Often the research interest in causal inference is on the regression causal effect, which is the mean difference in the potential outcomes conditional on the covariates. In this paper, we use sufficient dimension reduction to estimate a lower dimensional linear combination of the covariates that is sufficient to model the regression causal effect. Compared with the existing applications of sufficient dimension reduction in causal inference, our approaches are more efficient in reducing the dimensionality of covariates, and avoid estimating the individual outcome regressions. The proposed approaches can be used in three ways to assist modeling the regression causal effect: to conduct variable selection, to improve the estimation accuracy, and to detect the heterogeneity. Their usefulness are illustrated by both simulation studies and a real data example.
topic	central causal effect subspace conditional causal effect heterogeneity variable selection
url	https://doi.org/10.1515/jci-2018-0015
work_keys_str_mv	AT luowei learningheterogeneityincausalinferenceusingsufficientdimensionreduction AT wuwenbo learningheterogeneityincausalinferenceusingsufficientdimensionreduction AT zhuyeying learningheterogeneityincausalinferenceusingsufficientdimensionreduction
_version_	1717768427086544896

Learning Heterogeneity in Causal Inference Using Sufficient Dimension Reduction

Similar Items