Hi I need to replace all missing values in a dataset with respective mean. Find the sample i wrote. %macro replace_missing(dataset);       proc means data=&dataset mean ; ;             output out=t (drop=_type_ _freq_) mean=/autoname;     run;       proc transpose data=t out=tr;       run; data _null_;       set Tr end=last ;       call symput('mean'||left(_n_),col1);       if last then call symput('maxi', _n_);             /* x=left(_N_) */;             /* put _N_ maxi last x ; */; run; data &dataset;       set &dataset;       array num(*) _numeric_;             %do n=1 %to &maxi;                   if num(&n)=. then num(&n)= &&mean&n;             %end; run; %mend; %replace_missing(drop_missing); run; But this is not working.. Please help ... View more

I have a dataset 'test' as   var1 var2 var3 var4....varn   abcd M 22 1500...   qwerty M 32 5000...   asdf F 21 100...   tyuif F 33 6500..   jklmn M 45 800..   fghtn F 18 4000..   fgtn F 28 4050..   abcd M 22 1500..   abcde M 24 1500.. I need to get the output as for only numeric variables Header 1 Header 2 Header 3 Header 4 Header 5 Variable Min Max Mean Std Var3 Var4 ... Varn for only character variables Header 1 Header 2 Header 3 Header 4 Variable Mode Freqency of Mode No:of distinct varaibles var1 var2 .. varn ... View more

how to calculate the mode, frequency of mode and no:of distinct categories of all character variables in a dataset. Prefer proc univariate. ... View more

i want to write a macro with the mentioned requirement Input - A sas dataset with dependent binary variable and one predicted probability variable Output - A dataset with following Stats by Decile - No. of Obs, No. of events, % Obs, % events, Event Rate, Average Predicted Probability, Lift, Cumulative Lift, Cumulative %events ... View more