Please try the below code , I assumed that the xml data is in a txt file   filename sam '~/sample.txt'; proc import datafile=sam out=sample dbms=tab replace; getnames=no; run; data want; set sample; retain streetAndNumber; if prxmatch('m/streetAndNumber/oi',var1)>0 then streetAndNumber=prxchange('s/(.*>)(.*)(<.*)/$2/oi',-1,var1); if prxmatch('m/city/oi',var1)>0 then cityasaddress=prxchange('s/(.*>)(.*)(<.*)/$2/oi',-1,var1); QUALIFICATION_IDENTIFICATION='EETSTEE1000EE-2019-INR53390'; if cityasaddress ne ''; drop var1; run; ... View more

Please try the below code   check the want dataset where size has value 1 indicating that for id=3 there is difference in size   data one; input id size; cards; 1 3 2 4 3 5 4 6 5 . ; data two; input id size; cards; 2 4 3 6 4 6 5 7 ; proc compare base=one compare=two out=want listall; id id; run; ... View more

You can try something as below   data want; set have; flag_age=(a_age=b_age); flag_name=(m_name=l_name=c_name); run; ... View more

Could you please try the below code and I believe that you don't want to expand the number of rows per ID, but just need the count in days for every id and start date, so you get the count variable something like in the screenshot. Please explain if i did not understand your query.   data have; input Prod_Id Start_date:datetime18.6 End_date:datetime18.6 ; cards; 10644800 26FEB2020:04:05:50.180000 27FEB2020:04:25:54.189999 10656210 02SEP2019:04:15:36.840000 04SEP2019:04:03:47.819999 10656210 07MAR2020:06:25:11.780000 08MAR2020:06:37:29.919999 10656210 17FEB2020:03:55:08.130000 18FEB2020:04:58:05.809999 10665400 21AUG2019:12:44:09.780000 24AUG2019:05:16:01.809999 10665400 06NOV2019:05:51:53.420000 11NOV2019:05:14:50.279999 10682600 27FEB2020:04:25:54.190000 28FEB2020:04:25:13.529999 10703611 29SEP2019:04:04:38.320000 30SEP2019:05:19:45.929999 10703611 15DEC2019:04:12:49.910000 21DEC2019:04:40:18.419999 10703611 16NOV2019:05:19:58.700000 17NOV2019:04:43:56.199999 10703611 21AUG2019:12:44:09.780000 24AUG2019:05:16:01.809999 10703611 06NOV2019:05:51:53.420000 11NOV2019:05:14:50.279999 ; data want; set have; by Prod_Id Start_date notsorted; new_Start_date=datepart(Start_date); new_End_date=datepart(End_date); format new_Start_date new_End_date i date9.; do i=new_Start_date to new_End_date; count=(new_End_date-new_Start_date)+1; end; drop i; run; ... View more

Please try   new_cylinders=put(Cylinders , best.); ... View more

Please try the FILENAME statement option TERMSTR.    filename sample 'sample.csv' termstr=CRLF; proc import datafile=sample out=csv replace dbms=csv; run; ... View more

Please try the below code   data have; input ID$ Date:mmddyy10. FirstPositive; cards; A 1/1/2020 . A 1/3/2020 1 A 1/4/2020 . A 1/5/2020 . B 1/4/2020 1 B 1/5/2020 . ; data want; set have; by id notsorted; retain FirstPositive2; if first.id then do;FirstPositive2=.;count=.;end; if FirstPositive ne . then FirstPositive2=FirstPositive; count+FirstPositive2; if FirstPositive2 ne . then diff=Date-lag(date); if first.id then diff=.; run; ... View more

It is due to the way the data is sorted and location of variables , like in the below example  sorting by and variables order of member hospital admission will give 4 records due to hospital values distinguish the date 20160106 whereas in second example sorting by and variable order member admission hospital distinguish by admission date and we get 3 records where hospital is ignored as we are considering the admission date.       ... View more

Alternatively with ifn and retain   data want; set have; retain new; new=ifn(x<new,new,x); run; ... View more

please try below code   data T; input ID $ L M; Datalines; B 1 10 B 1 20 C 1 30 C 2 40 D 1 50 D 1 60 D 2 70 ; proc sql; create table want as select ID, L, M, sum(M) as sumM from T group by ID, L ; quit; ... View more