2009년 10회 통계세미나 개최 안내
통계연구소에서는 다음과 같이 통계 세미나를 개최하오니 많은 참여 바랍니다.
일시: 2009년 11월 25일(수) 오후 5시
장소: 고려대학교 정경관 507호
연사: Moonsu Kang
(Department of Medicine, Hanyang university)
"Power and False Discovery Rate in Genomic
Studies: A Two-Stage Procedure in Perspectives."
In microarray and other genomic studies, in view of an abundance of genes, one statistical approach is to hold the family wise error rate to a prescribed limit while minimizing the false discovery rate by suitable multiple hypothesis test-ing procedures, thus generally compromising the power properties to a certain extent. Power considerations are no less important in this context. Since the genes are not generally independent or even marginally identically distributed, model flexibility is an essential task in enhancing power perspectives. In this respect, incorporating a version of the Chen-Stein theorem along with some novel bivariate exceedance probabilities, at high level of crossing, some two- or multi-stage procedures have been considered; they seem to have better power without much elevation of false discovery rate. Numerical studies are also ap-pended with the methodological developments.
Keywords: Microarray data; Familywise error rate; False discovery rate;Chen-Stein method; Discrete p-values; Bivariate extremes
고려대학교 통계연구소