Problems of repeated significance testing.
暂无分享,去创建一个
Abstract Repeated significance testing at several time points and for more than one variable causes the inflation of the probability for the error of the first kind, i.e., the probability of rejecting the global null hypothesis of no differences between the true treatment effects when in fact this hypothesis is true. Three alternatives to the generally overconservative Bonferroni procedure are proposed where each of these alternatives may suit different conditions as set by the physician.
[1] D. Morgenstern,et al. Berechnung des maximalen Signifikanzniveaus des Testes „LehneH0 ab, wennk untern gegebenen Tests zur Ablehnung führen” , 1980 .
[2] William Feller,et al. An Introduction to Probability Theory and Its Applications , 1951 .
[3] B. Rüger. Das maximale signifikanzniveau des Tests: “LehneHo ab, wennk untern gegebenen tests zur ablehnung führen” , 1978 .