I liked @A_Kh's suggestion to sort by USUBJID descending ACTARM, which is intended to put missing ACTARM obs last within a USUBJID.  But without a minor edit, it would not accommodate two distinct valid values for ACTARM (if a USUBJID had a valid "B" and a valid "C", the "B" would be deleted).   Consider this alternative, applied to data HAVE sorted by USUBJID (but not necessarily by ACTARM within USUBJID):   data want; set have (where=(missing(actarm)=0) in=valid) have (where=(missing(actarm)) in=miss); by usubjid; if lag(valid)=1 and first.usubjid=0 and miss=1 then delete; run;   ... View more

Here is a code to create the arrays you need. I leave it to you to add the computation. Data is assumed to be sorted by subject; data computed; set have; by subject; retain ndisp1-ndisp4 pil1-pil4; array nd {4} ndisp1-ndisp4; array pl {4} pil1-pil4; if first.subject then do i=1 to 4; /* clear arrays */ nd(i)=.; pl(i)=.; end; /* do for every observation per subject = save data in the arrays */ i = input(substr(week,1,1),1.); /* week nomber 1 to 4 */ nd(i) = numdisp; pl(i) = pilre; if last.subject then do; /* compute the CDOSE by formula */ ...... output; end; keep subject cdose; run;  You may need to merge HAVE and COMPUTED tables by SUBJECT to have CDOSE variable on each observation. ... View more

data chk; input id var1$ var2$ var3$ var4$ var5$; array v(*) var:; array x(5) $ 8; n=0; do i=1 to dim(x); if v[i] not in x then do;n+1;x[n]=v[i];end; end; var6 = catx(',',of x(*)); drop i x: n; datalines; 1 mic advil keyboard advil mouse 2 mouse mouse keyboard perfume notebook 3 book wire plug desktop charger 4 mobile laptop mouse desktop phone ; run; ... View more

Thank you for your quick response. Used this code and worked perfectly ... View more