Thanks for the information I will look into that blog... Is it possible to create Sub-grouped donut chart for multiple level nested data ? ... View more

Yes Exactly... But I dont have access to SAS VA.. So is there any other way to do this ? ... View more

Use proc transpose.. proc transpose data=have out=want (drop=_name_); by Questions notsorted; id Dummy; var Score; run; ... View more

Hi.. I think this might work for you.. Data have; infile datalines missover; input x1 x2 $ x3 $ x4 $ products :$100.; datalines4; 1  a   a   a  car;tv;fridge;washingmachine;mobile           2   b   b  b Car;fridge;phone     ;;;; run; proc sql noprint; select put((max(count(products,";"))+1),best.) into :tot from have; quit; data want(drop=i); set have; array pro(*) $30 product1-product%sysfunc(strip(&tot)); do i=1 to %sysfunc(strip(&tot)); pro(i)=scan(products,i,";"); end; run; ... View more

Hi Zana, Considering the case as you have mentioned below: fortunately the IDs in this data set can’t have different year. Each ID will have distinctly one year only, In this scenario below code will help for you: PROC SQL; CREATE TABLE WANT AS   SELECT A.ID, OBS, YEAR FROM ONE AS A INNER JOIN       (SELECT ID FROM (SELECT DISTINCT ID, YEAR FROM ONE)        GROUP BY YEAR HAVING COUNT(YEAR) GT 2) AS B     ON A.ID=B.ID; QUIT; ... View more

Try now PROC SQL; CREATE TABLE WANT AS SELECT *, CASE WHEN COND1=. OR COND2=. THEN "MISSING" WHEN COND1 LE 0 AND COND2 LE 0 AND COND3 =" " THEN "YES" WHEN COND1 GT 0 AND COND2 GT 0 AND COND3= " " THEN "NO" /*ELSE "UNKNOWN"*/ END AS NEWVAR LENGTH=10 FROM HAVE; QUIT; Message was edited by: SK K ... View more

Try this , it works fine for ur condition and i have also added another value "Unknown" for data which does not meet any of the conditions PROC SQL; CREATE TABLE WANT AS SELECT *, CASE WHEN COND1=. AND COND2=. THEN "MISSING" WHEN COND1 LE 0 AND COND2 LE 0 AND COND3 =" " THEN "YES" WHEN COND1 GT 0 AND COND2 GT 0 AND COND3= " " THEN "NO" /*ELSE "UNKNOWN"*/ END AS NEWVAR LENGTH=10 FROM HAVE; QUIT; ... View more

Hi.. Here's something with Proc SQl.. I manipulated your input data a little bit.. to get expected output.., data have; infile datalines missover; input sno cond1 cond2 cond3 $; datalines; 1              4          8               2              .          .     3              -1          8 4              .          1               min 5             -1           -2    ; run; PROC SQL; CREATE TABLE WANT AS SELECT *, CASE WHEN COND1=. AND COND2=. THEN "MISSING" WHEN COND1 LT 0 AND COND2 LT 0 AND COND3 =" " THEN "YES" WHEN COND1 GT 0 AND COND2 GT 0 AND COND3= " " THEN "NO" ELSE "UNKNOWN" END AS NEWVAR LENGTH=10 FROM HAVE; QUIT; ... View more