Hello,   You have to sort your data first    or   use Notsorted option with By statement.      by HCN REG_TM notsorted;     ... View more

you can import your excel file by using PROC IMPORT procedure.   just need to update filepath in %let path statement.    %let path=filepath; proc import datafile="&path.\data.xlsx" out=want dbms=excel replace; run; ... View more

data want; length newvar $5.; input (v1-v5) ($) ; call catt(newvar,substr(v1,1,1),substr(v2,1,1),substr(v3,1,1),substr(v4,1,1),substr(v5,1,1)); cards ; john peter max jann kiyaa ; run; ... View more

Hello,   In your case, you can use a blank to represent a missing value for character variable .   For you example , in place of .(dot) if you want to place a space (blank) you can use DSD and DLM options with infile statement.    data tbl; length name $ 12; infile datalines dsd dlm=''; input name $ score1 score2; datalines; Riley 1132 1187 Henderson 1015 1102 1018 1027 ; Run; ... View more

Hi Allaluiah, I tried to solve the problem in other way. Thank you for your comment. Here is the simple code .. proc sql; create table want as select subject_id ,visit_day ,expected_visit_date,actual_visit_date , case when expected_visit_date <= max(actual_visit_date) then coalesce(actual_visit_date,expected_visit_date) else . end as new_date format=mmddyy. from have group by subject_id order by subject_id,visit_day; quit; ... View more

data have; input subject_id visit_day expected_visit_date : mmddyy10. actual_visit_date mmddyy10.; format expected_visit_date actual_visit_date mmddyy10.; datalines; 1 1 1/1/2017 1/1/2017 1 2 1/2/2017 1/2/2017 1 3 1/3/2017 1 4 1/4/2017 1/4/2017 2 1 2/1/2017 2/1/2017 2 2 2/2/2017 2 3 2/3/2017 2 4 2/4/2017 2/4/2017 2 5 2/5/2017 2 6 2/6/2017 2/6/2017 3 1 3/1/2017 3/1/2017 3 2 3/2/2017 3/2/2017 3 3 3/3/2017 3/3/2017 3 4 3/4/2017 3/4/2017 3 5 3/5/2017 3/5/2017 3 6 3/6/2017 ; run; proc sql; create table want as select Subject_id,visit_day,expected_visit_date,coalesce(actual_visit_date,newd) as actual_visit_date format=mmddyy10. from (select Subject_id,visit_day,expected_visit_date,actual_visit_date,case when max(actual_visit_date) gt 0 and (v gt visit_day) then coalesce(actual_visit_date,expected_visit_date) else . end as newd format=mmddyy10. from (select Subject_id, visit_day , expected_visit_date,actual_visit_date,max(visit_day) as v from have group by Subject_id)) order by subject_id,visit_day; quit; ... View more

Hope so below Code will match your exact requirement ...   Not able to copy and paste .. so posted as screen shot....         ... View more

Hello,   below code may be helpful ..   proc sql; create table want as select distinct Status,COL1,TREATMENT_A1,TREATMENT_B1,sum(TREATMENT_A1,TREATMENT_B1) as total from( select 'Responders' as Status, 'N' as COL1,id,trt01pn, case when trt01pn=1 then count(*) end as TREATMENT_A1, case when trt01pn=2 then count(*) end as TREATMENT_B1 from have group by trt01pn); run; proc print data=want; sum total; run; ... View more