Introduction
Methods
(1)Bonferronni校正
(2)Sidak校正
α’ 是開頭講的1類累計誤差;則公式可以變形爲:
α=1-(1-α’)^1/c
這方法用在當確定了我們需要的顯著性水平α後,來求實際的α值,將我們確定的α帶入α’中,如前面的例子α=0.05,帶入α’中,求得1-(1-0.05)^1/10 =0.005;則實際我們用的α值應設爲0.00512619…,也就約等於0.005。和Bonferonni的結果近似,0.05/10=0.005 。但是α=1-(1-α’)^1/c ≧ α’/c,這裏也看出了Bonferonni校正過於保守。
(3)Hochberg校正
將原始P 值按照大小排序:p(1) ≦ p(2)… ≦ p(k) ≦p(m);k從1到m;p(k) ≦α/ (m - k+ 1);
the largest k ;
《From: Hochberg, Yosef (1988). “A Sharper Bonferroni Procedure for Multiple Tests of Significance”》
(4)FDR校正
(5)BH校正——加權
Weighted versionLet p1,..., pk be the unadjusted p-values and let w1,..., wk be a set of corresponding positive weights that add to 1. Without loss of generality, assume the p-values and the weights are all ordered such that p1/w1 ≤ p2/w2 ≤ ... ≤ pk/wk.The adjusted p-value for the first hypothesis is q1 = min{1,p1/w1}. Inductively, define the adjusted p-value for hypothesis i by qi = min{1,max{qi−1,(wi + ... + wk)×pi/wi}}.