Need help in calculating stratified p-value in my pgm

[knveraraju91]

Dear

Please suggest me in my pgm how to calculate stratified p_value. The stratification factors are (age ss ph). Thank you

data one;
input trt res$ age ss ph $;
datalines;
1 cr 55 1 Y
1 cr 53 1 N
1 cri 52 1 Y
1 cri 58 1 N
1 cri 50 2 N
1 cr 53 2 N
1 cri 52 2 Y
1 cri 58 2 N
2 cri 56 2 Y
2 cri 51 2 N
2 cr 50 2 Y
2 cr 56 2 N
;

  proc freq data = one;
                                          table age*ss*ph*trt*res/ alpha = 0.05 chisq pdiff cl nocol nopercent ;
                                          exact chisq / alpha = 0.05 ;
                                          ods output chisq = exact (where = (upcase(statistic) = 'CHI-SQUARE')) ;
                                          output out = pval(rename = (p_pchi = pval)) pchi ;
                                  run ;

[Ksharp]

/*If it is high dimension contingency table ,I think you need calculated CMH statistic*/
proc freq data = one;
table age*ss*ph*trt*res/ alpha = 0.05 cmh ;
run ;


And better post it at Statistical Forum :
https://communities.sas.com/t5/Statistical-Procedures/bd-p/statistical_procedures

[ballardw]

Are you requesting stratification by Age (only) SS (only) and PH (only) for the TRT*RES? If so perhaps

proc freq data = one;
   table (age ss ph)*trt*res/ alpha = 0.05 chisq pdiff cl nocol nopercent ;
   exact chisq / alpha = 0.05 ;
   ods output chisq = exact (where = (upcase(statistic) = 'CHI-SQUARE')) ;
   output out = pval(rename = (p_pchi = pval)) pchi ;
run ;

You can group variables by role in Proc Freq table statements with ( )