Thank you, Dan. ALT= works! I had been trying with Title= for some time. Whilst it worked in a stored process, it didn't function correctly in a web report. ALT works fine. Thanks. ... View more

Thank you! ... View more

Thank you, Jag! It works well. VD ... View more

Times 100. title "276/(276+543)=%sysevalF(276/(276+543)) for NBUS, 264/(264+1099)=%sysevalF(264/(264+1099)) for Season"; proc freq data=have;    tables status*tenure / norow nopercent;    weight count;    run; ... View more

Thank you everyone for the help. This is perfect =). ... View more

The HASH object allows you to create a dynamic array indexed on CUSTID and accumulate the counts. data email;    infile datalines eof=eof;    input custid $ date MMDDYY8. response $;    length count 8;    if _n_ eq 1 then do;       declare hash h(ordered:'Y');       h.definekey('custid');       h.definedata('custid','count');       h.definedone();       end;    call missing(count);    rc=h.find();    count + 1;    rc=h.replace();    return; eof:    h.output(dataset:'count');    stop;    format date MMDDYY10.;    drop count rc;    datalines; A 12-22-10 Y B 09-18-07 N C 08-11-11 Y A 03-19-03 Y B 09-01-01 N C 08-03-12 N D 03-01-04 Y A 11-08-08 Y B 03-09-09 Y C 09-06-09 Y E 09-08-09 Y A 09-09-09 Y C 09-06-09 Y B 09-01-06 N E 09-03-12 Y D 09-23-11 Y A 01-01-00 Y C 09-08-02 Y D 09-07-00 Y F 09-09-09 Y B 09-01-00 N C 01-01-00 N ; run; ... View more

Ahh I was not aware, but this is very useful information. Thank you!! ... View more

Hi Data _null_ You understand the question right. And you gave the right one. Thanks for your help. i am using lenthy code using retain statment, because date one have few records on trigdt and data two have few records on sampdt . But i need both record in my final dataset.Now i got it with your help Thanks Sam ... View more