Multiple Hypothesis Test Error Rate Control Analysis Of The Process

The use of the multiple hypothesis testing has been receiving a lot of attention recently by analying large-dimension data.In this condition,we need to test thousands of hypotheses simultaneously.When simultaneously undertaking many tests of signifiicance researchers are faced with the problem of how best to control the probability of commiting a TypeⅠerror.The multiple hypothesis testing problem has long been takled by controlling the family-wise error rate(FWER),the probability of even one false rejection. Clearly,the traditional FWER controlling procedures are too stringent.When the number of hypotheses is large,researchers might be willing to tolerate more than one false rejection,thereby increasing the ability of the procedure to detect false null hypotheses. Consequently,it is desirable to consider other new measure.Benjamini and Hochberg (1995)proposed FDR,at the same time they provided BH procedure for controlling FDR. FWER and FDR are important in multiple hypothesis testing area,many reseachers are studying them.So,in the first part of this study,we review definitions and procedures of FWER and FDR.In the second part of the study,we propose a new method to estimate the number of true null hypotheses.The estimates can be used in FDR-controlling procedure with a consequent increase in power.