Variable selection in logistic regression for detecting SNP-SNP interactions: the rheumatoid arthritis example.

ABSTRACT: Many complex disease traits are observed to be associated with single nucleotide polymorphism (SNP) interactions. In testing small-scale SNP-SNP interactions, variable selection procedures in logistic regressions are commonly used. The empirical evidence of variable selection for testing interactions in logistic regressions is limited. This simulation study was designed to compare nine variable selection procedures in logistic regressions for testing SNP-SNP interactions. Data on 10 SNPs were simulated for 400 and 1000 subjects (case/control ratio=1). The simulated model included one main effect and two 2-way interactions. The variable selection procedures included automatic selection (stepwise, forward and backward), common 2-step selection, AIC- and SC-based selection. The hierarchical r

SUBMITTER: Lin HY 

PROVIDER: S-EPMC3786179 | biostudies-literature | 2008 Jun

REPOSITORIES: biostudies-literature