A BINOMIAL MODEL APPROXIMATION FOR MULTIPLE TESTING

Abstract

Multiple testing is associated with simultaneous testing of many hypotheses, and frequently calls for adjusting level of significance in some way that the probability of observing at least one significant result due to chance remains below the desired significance levels. This study developed a Binomial Model Approximations (BMA) method as an alternative to addressing the multiplicity problem associated with testing more than one hypothesis at a time. The proposed method has demonstrated capacity for controlling Type I Error Rate as sample size increases when compared with the existing Bonferroni and False Discovery Rate (FDR).