TY - JOUR

T1 - Credsubs

T2 - Multiplicity-adjusted subset identification

AU - Schnell, Patrick M.

AU - Fiecas, Mark

AU - Carlin, Bradley P.

N1 - Funding Information:
This work was supported by AbbVie, Inc.

PY - 2020

Y1 - 2020

N2 - Subset identification methods are used to select the subset of a covariate space over which the conditional distribution of a response has certain properties – for example, identifying types of patients whose conditional treatment effect is positive. An often critical requirement of subset identification methods is multiplicity control, by which the family-wise Type I error rate is controlled, rather than the Type I error rate of each covariate-determined hypothesis separately. The credible subset (or credible subgroup) method provides a multiplicity-controlled estimate of the target subset in the form of two bounding subsets: one which entirely contains the target subset, and one which is entirely contained by it. We introduce a new R package, credsubs, which constructs credible subset estimates using a sample from the joint posterior distribution of any regression model, a description of the covariate space, and a function mapping the parameters and covariates to the subset criterion. We demonstrate parametric and nonparametric applications of the package to a clinical trial dataset and a neuroimaging dataset, respectively.

AB - Subset identification methods are used to select the subset of a covariate space over which the conditional distribution of a response has certain properties – for example, identifying types of patients whose conditional treatment effect is positive. An often critical requirement of subset identification methods is multiplicity control, by which the family-wise Type I error rate is controlled, rather than the Type I error rate of each covariate-determined hypothesis separately. The credible subset (or credible subgroup) method provides a multiplicity-controlled estimate of the target subset in the form of two bounding subsets: one which entirely contains the target subset, and one which is entirely contained by it. We introduce a new R package, credsubs, which constructs credible subset estimates using a sample from the joint posterior distribution of any regression model, a description of the covariate space, and a function mapping the parameters and covariates to the subset criterion. We demonstrate parametric and nonparametric applications of the package to a clinical trial dataset and a neuroimaging dataset, respectively.

KW - Credible subgroups

KW - Multiple hypothesis testing

KW - R, Subset identification

KW - Subgroup analysis

UR - http://www.scopus.com/inward/record.url?scp=85090518091&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85090518091&partnerID=8YFLogxK

U2 - 10.18637/jss.v094.i07

DO - 10.18637/jss.v094.i07

M3 - Article

AN - SCOPUS:85090518091

VL - 94

SP - 1

EP - 22

JO - Journal of Statistical Software

JF - Journal of Statistical Software

SN - 1548-7660

ER -