Example 43.5: Inputting Raw p-Values

The MULTTEST Procedure

Example 43.5: Inputting Raw p-Values

This example illustrates how to use PROC MULTTEST to multiplicity-adjust a collection of raw p-values obtained from some other source. This is a valuable option for those cases where PROC MULTTEST cannot compute the raw p-values directly.

   data a;
       input Test$ Raw_P;
       datalines;
   test1 .09108
   test2 .69122
   test3 .00177
   test4 .57181
   test5 .03121
   test6 .01413
   ;

   proc multtest pdata=a holm hoc fdr out=new;
   run;

   proc print data=new;
   run;

Note that there are no statements other than the PROC MULTTEST statement using the p-value input mode. In this example, the raw p-values are adjusted using the Holm, Hochberg, and Benjamini and Hocherg (FDR) methods. The OUT= data set specification is required. PROC MULTTEST produces no output other than this output data set in this case, and resampling-based adjusted p-values cannot be computed.

The OUT= data set from this analysis is listed in Output 43.5.1.

Output 43.5.1: Inputting Raw p-Values

Obs	Test	Raw_P	stpbon_p	hoc_p	fdr_p
1	test1	0.09108	0.27324	0.27324	0.13662
2	test2	0.69122	1.00000	0.69122	0.69122
3	test3	0.00177	0.01062	0.01062	0.01062
4	test4	0.57181	1.00000	0.69122	0.68617
5	test5	0.03121	0.12484	0.12484	0.06242
6	test6	0.01413	0.07065	0.07065	0.04239

Note that the adjusted p-values for the Hochberg method (hoc_p) are less than or equal to those for the Holm method (stpbon_p). In turn, the adjusted p-values for the Benjamini and Hochberg method (fdr_p) are less than or equal to those for the Hochberg method. These comparisons hold generally for all p-value configurations. The FDR method controls the false discovery rate and not the familywise error rate. The Hochberg method controls the familywise error rate under independence. The Holm method controls the familywise error rate without assuming independence.

Chapter Contents
Previous
Next
Top