How to relate potential outcomes: Estimating individual treatment effects under a given specified partial correlation

Mingyang Cai; Stef van Buuren; Gerko Vink

doi:10.48550/arXiv.2208.12931

How to relate potential outcomes: Estimating individual treatment effects under a given specified partial correlation

Mingyang Cai
, Stef van Buuren
, Gerko Vink

Research output: Working paper › Preprint › Academic

Abstract

In most medical research, the average treatment effect is used to evaluate a treatment’s performance. However, precision medicine requires knowledge of individual treatment effects: What is the difference between a unit’s measurement under treatment and control conditions? In most treatment effect studies, such answers are not possible because the outcomes under both experimental conditions are not jointly observed. This makes the problem of causal inference a missing data problem. We propose to solve this problem by …

Original language	English
Publisher	arXiv
Pages	1-37
DOIs	https://doi.org/10.48550/arXiv.2208.12931
Publication status	Published - 2022

Keywords

Multiple imputation
joint modeling imputation
iterativeimputation
multivariate data analysis

Access to Document

10.48550/arXiv.2208.12931Licence: CC BY

2208.12931Submitted manuscript, 398 KBLicence: CC BY

Cite this

@techreport{d03dcb2222734744b4ee5a5bafa74897,

title = "How to relate potential outcomes: Estimating individual treatment effects under a given specified partial correlation",

abstract = "In most medical research, the average treatment effect is used to evaluate a treatment{\textquoteright}s performance. However, precision medicine requires knowledge of individual treatment effects: What is the difference between a unit{\textquoteright}s measurement under treatment and control conditions? In most treatment effect studies, such answers are not possible because the outcomes under both experimental conditions are not jointly observed. This makes the problem of causal inference a missing data problem. We propose to solve this problem by …",

keywords = "Multiple imputation, joint modeling imputation, iterativeimputation, multivariate data analysis",

author = "Mingyang Cai and \{van Buuren\}, Stef and Gerko Vink",

year = "2022",

doi = "10.48550/arXiv.2208.12931",

language = "English",

pages = "1--37",

publisher = "arXiv",

type = "WorkingPaper",

institution = "arXiv",

}

TY - UNPB

T1 - How to relate potential outcomes: Estimating individual treatment effects under a given specified partial correlation

AU - Cai, Mingyang

AU - van Buuren, Stef

AU - Vink, Gerko

PY - 2022

Y1 - 2022

N2 - In most medical research, the average treatment effect is used to evaluate a treatment’s performance. However, precision medicine requires knowledge of individual treatment effects: What is the difference between a unit’s measurement under treatment and control conditions? In most treatment effect studies, such answers are not possible because the outcomes under both experimental conditions are not jointly observed. This makes the problem of causal inference a missing data problem. We propose to solve this problem by …

AB - In most medical research, the average treatment effect is used to evaluate a treatment’s performance. However, precision medicine requires knowledge of individual treatment effects: What is the difference between a unit’s measurement under treatment and control conditions? In most treatment effect studies, such answers are not possible because the outcomes under both experimental conditions are not jointly observed. This makes the problem of causal inference a missing data problem. We propose to solve this problem by …

KW - Multiple imputation

KW - joint modeling imputation

KW - iterativeimputation

KW - multivariate data analysis

U2 - 10.48550/arXiv.2208.12931

DO - 10.48550/arXiv.2208.12931

M3 - Preprint

SP - 1

EP - 37

BT - How to relate potential outcomes: Estimating individual treatment effects under a given specified partial correlation

PB - arXiv

ER -

How to relate potential outcomes: Estimating individual treatment effects under a given specified partial correlation

Abstract

Keywords

Access to Document

Fingerprint

Cite this