Single-Marker Analysis for Unmatched Case-Control Data

Abstract

AbstractChapter 3 begins with an introduction to the notation for penetrance and genotype relative risk. Since many statistical analyses of a case-control association study depend on the underlying genetic model, the genetic models are introduced in terms of genotype relative risks. Test statistics for genetic association covered in this chapter include genotype-based tests (Pearson’s chi-squared test, the Cochran-Armitage trend test, and the likelihood-ratio test), the allelic test, exact tests, and the Hardy-Weinberg disequilibrium trend test. Combining the Hardy-Weinberg disequilibrium trend test with the Cochran-Armitage trend test is also discussed. Numerical and analytical comparisons between the allelic and genotype-based tests are given. This chapter also discusses how to obtain the trend test and Pearson’s chi-squared test as the Score tests from logistic regression models. Estimates of odds ratios and their confidence intervals are given. Results from simulation studies are presented. Common approaches to simulate case-control data with or without covariates are given. Examples and case studies are presented.KeywordsCase-control SampleCochran-Armitage Trend Test (CATT)Common Genetic ModelsGenotype CountsParametric Bootstrap MethodThese keywords were added by machine and not by the authors. This process is experimental and the keywords may be updated as the learning algorithm improves.