Robust statistical tests of genetic association for the case- control study design

1 The central theme in case-control genetic association studies is to efficiently identify genetic markers associated with case-control status. Powerful statistical methods are critical to accomplishing this goal. A popular statistical method is the model-free Pearson's chi-square test. To achieve increased power, model-based tests have been widely used despite their lack of robustness to model-misspecification. Much research has been carried out on increasing the robustness of model-based tests. A model-free analysis framework is proposed. It involves less degree of freedom than the Pearson's chi-square test. The likelihood ratio statistic, the score statistic, and the Wald statistic are introduced. In addition, these statistics are less affected by the confounding effect of population stratification on genetic association. The performance of these statistics are evaluated by computer simulation. Also introduced is a test for the existence of population stratification. This statistic is asymptot-ically uncorrelated with the proposed lieklkihood ratio statistic, the score statistic, and the Wald statistic. All these statistics are applied to a study of height in a European American population.