how about linear count with Hash table? data One; input ID $12. Var1 Var2 Var3; cards; 111-111-1111 0 1 0 222-222-2222 0 1 1 333-333-3333 1 1 1 444-444-4444 0 1 0 555-555-5555 1 1 0 141-547-7810 1 1 0 ; run; proc print data=One; run; Data Two; input ID $12. Var1 Var2 Var3; cards; 111-111-1111 0 1 0 666-666-6666 0 1 1 333-333-3333 0 1 1 444-444-4444 0 1 0 555-555-5555 0 1 0 777-777-8888 0 0 0 ; run; proc print data=Two; run; data _null_; dcl hash H(ordered:"A"); H.defineKey("ID"); H.defineData("ID","inONE","inTWO"); H.defineDone(); do until(_E_); set ONE(keep=ID in=in1) TWO(keep=ID in=in2) end=_E_; rc=H.find(); inONE+in1; inTWO+in2; /*put _ALL_;*/ /* just for preview */ rc=H.replace(); call missing(inONE,inTWO); end; h.Output(dataset:"unique"); stop; run; proc print data=unique; run; Bart ... View more

  @mkeintz  Thing is that DS2 doesn't support WHERE filtering... 1 proc ds2; 2 data want(overwrite=yes); 3 method run(); 4 set have; 5 where age=16; 6 end; 7 enddata; 8 run; 8 ! quit; ERROR: Compilation error. ERROR: Missing END statement for the method run. ERROR: Parse encountered WHERE when expecting end of input. ERROR: Line 5: Parse failed: >>> where <<< age=16; NOTE: The SAS System stopped processing this step because of errors. NOTE: PROCEDURE DS2 used (Total process time): real time 0.01 seconds user cpu time 0.01 seconds system cpu time 0.00 seconds memory 3864.96k OS Memory 21752.00k We have to use {SQL subqueries} 1 proc ds2; 2 data want(overwrite=yes); 3 method run(); 4 set { select * from have1 where age=16 }; 5 end; 6 enddata; 7 run; 7 ! quit; NOTE: Execution succeeded. 9 rows affected. NOTE: PROCEDURE DS2 used (Total process time): real time 0.03 seconds user cpu time 0.03 seconds system cpu time 0.00 seconds memory 4628.15k OS Memory 22512.00k @Ksharp  I did small experiment, on 10M observations: data have1 have2 have3 have4; call streaminit(123); do i=1 to 1e7; age=rand('integer',1,100000);output; end; run; Data Step WHERE seems to be winning (at least at my laptop) 10 11 12 proc ds2; 13 data want1(overwrite=yes); 14 method run(); 15 set have1 ; 16 if age=16; 17 end; 18 enddata; 19 run; 19 ! quit; NOTE: Execution succeeded. 96 rows affected. NOTE: PROCEDURE DS2 used (Total process time): real time 0.13 seconds user cpu time 0.17 seconds system cpu time 0.00 seconds real time 0.12 seconds user cpu time 0.17 seconds system cpu time 0.03 seconds real time 0.13 seconds user cpu time 0.12 seconds system cpu time 0.04 seconds real time 0.13 seconds user cpu time 0.14 seconds system cpu time 0.03 seconds real time 0.12 seconds user cpu time 0.12 seconds system cpu time 0.07 seconds memory 4648.96k OS Memory 22256.00k 20 21 proc ds2; 22 data want2(overwrite=yes); 23 method run(); 24 set { select * from have2 where age=16 }; 25 end; 26 enddata; 27 run; 27 ! quit; NOTE: Execution succeeded. 96 rows affected. NOTE: PROCEDURE DS2 used (Total process time): real time 0.13 seconds user cpu time 0.11 seconds system cpu time 0.03 seconds real time 0.12 seconds user cpu time 0.11 seconds system cpu time 0.01 seconds real time 0.14 seconds user cpu time 0.12 seconds system cpu time 0.01 seconds real time 0.12 seconds user cpu time 0.09 seconds system cpu time 0.03 seconds real time 0.12 seconds user cpu time 0.11 seconds system cpu time 0.01 seconds memory 4628.23k OS Memory 22512.00k 28 29 data want3; 30 set have3 ; 31 if age=16; 32 run; NOTE: There were 10000000 observations read from the data set WORK.HAVE3. NOTE: The data set WORK.WANT3 has 96 observations and 2 variables. NOTE: DATA statement used (Total process time): real time 0.16 seconds user cpu time 0.12 seconds system cpu time 0.04 seconds real time 0.16 seconds user cpu time 0.17 seconds system cpu time 0.00 seconds real time 0.15 seconds user cpu time 0.14 seconds system cpu time 0.01 seconds real time 0.15 seconds user cpu time 0.14 seconds system cpu time 0.00 seconds real time 0.14 seconds user cpu time 0.10 seconds system cpu time 0.03 seconds memory 624.21k OS Memory 20464.00k 33 34 data want4; 35 set have4 ; 36 where age=16; 37 run; NOTE: There were 96 observations read from the data set WORK.HAVE4. WHERE age=16; NOTE: The data set WORK.WANT4 has 96 observations and 2 variables. NOTE: DATA statement used (Total process time): real time 0.10 seconds user cpu time 0.09 seconds system cpu time 0.01 seconds real time 0.10 seconds user cpu time 0.09 seconds system cpu time 0.00 seconds real time 0.10 seconds user cpu time 0.09 seconds system cpu time 0.01 seconds real time 0.10 seconds user cpu time 0.07 seconds system cpu time 0.03 seconds real time 0.10 seconds user cpu time 0.07 seconds system cpu time 0.01 seconds memory 641.87k OS Memory 20464.00k   Bart     ... View more

/* fake data */ libname b (work); data short b.rti(keep=id sflag); id = 1; sflag = 2; date = "06may2024"d; days = 42; m_score = 0.16; run; proc sql; create table work.short2 as select a.*, case when missing(a.date) then . when a.date<="05may2024"d then case when a.m_score >= 0.13 then 1 else 0 end else case when a.date >="06may2024"d then case when a.Days >= 20 then 1 else 0 end ELSE . /* missing ELSE */ end end as T_Flag from short a left join b.rti b on a.id = b.id and a.sflag = b.sflag ; quit; ... View more

Some fun with macro looping could work too: %macro loopFormat(s,e,b=1,r=1,z=00); %do i=&s. %to &e. %by &b.; %eval(&i.-&b.)&z.<-&i.&z.="(%eval(&i.+&r.)) %eval(&i.-&b.)&z. - &i.&z." %end; %mend loopFormat; options Mprint; proc format; value F_Ronein 0='(1) 0' %loopFormat(1,5,r=1) %loopFormat(10,10,b=5,r=-3) %loopFormat(2,10,r=6,z=000) 10000<-15000='(17) 10000-15000' 15000<-High='(18) 15000+' ; Run; But @LinusH idea of input data set seems to be more practical. B. ... View more

It's called Maxim 4, you should start to use it.   [EDIT:] I agree with @LinusH adding an index could be good idea.   B. ... View more

One more for not sorted groups of families, see the data.   /* HASH TABLES - data not sorted family groups */ Data have; Input month GroupID IndividualID; GROUP_Id_MONTH=catx('-',GroupID,month); cards; 202401 123456 111 202401 123456 222 202401 123456 333 202401 123457 1111 202401 123457 2222 202401 123457 3333 202402 135798 555 202402 135798 666 202402 123457 1111 202402 123457 2222 202402 123456 111 202402 123456 222 202402 123456 444 202401 135798 555 202401 135798 666 ; Run; data _null_; if 1=_N_ then do; declare hash family(ordered:"A"); /* create first family setup in the first period */ family.defineKey('GroupID', 'IndividualID'); family.defineDone(); declare hash grpIDflag(); /* store flag for group */ grpIDflag.defineKey('GroupID'); grpIDflag.defineData('GroupID', 'flag'); grpIDflag.defineDone(); end; do fs = 1 by 1 until(last.GroupID); set have end=_E_; by month GroupID notsorted; if grpIDflag.find() then family.add(); /* add an individual to the first family */ else /* compare family setup in other periods */ flag + (family.check()) /* check for new member */ + (family.num_items NE fs)*last.month /* size change (e.g., from 3 to 2) */ ; end; flag = flag&1; grpIDflag.replace(); if _E_; grpIDflag.output(dataset:'want3'); /* family.output(dataset:'firstFamily'); */ keep GroupID flag; run; proc print data=want3; run;   Bart ... View more

One more approach with hash tables, I've added third group (123457) for tests.   Data have; Input month GroupID IndividualID; GROUP_Id_MONTH=catx('-',GroupID,month); cards; 202401 123456 111 202401 123456 222 202401 123456 333 202402 123456 111 202402 123456 222 202402 123456 444 202401 135798 555 202401 135798 666 202402 135798 555 202402 135798 666 202401 123457 111 202401 123457 222 202401 123457 333 202402 123457 111 202402 123457 222 ; Run; data want; declare hash family(); family.defineKey('IndividualID'); family.defineDone(); family.clear(); call missing(flag, period, fs); do until(last.GroupID); set have; by GroupID month notsorted; period+first.month; if 1=period then /* create first family setup in the first period */ family.add(); /* add an individual to the first family */ else /* compare family setup in other periods */ do; if first.month then fs=1; else fs+1; /* other family size */ flag + (family.check()) /* check for new member */ + (family.num_items NE fs)*last.month /* size change (e.g., from 3 to 2) */ ; end; end; flag = flag&1; output; keep GroupID flag; run; proc print; run; Bart   [EDIT:] And of course with arrays: data want2; array ids[100] _temporary_; /* 100 - assumed max family size */ call missing(flag, of ids[*], period, ffs, ofs); do until(last.GroupID); set have; by GroupID month notsorted; period+first.month; if 1=period then /* create first family setup in the first period */ do; ffs+1; /* first family size */ ids[ffs]=IndividualID; /* add individual to the first family */ end; else /* compare family setup in other periods */ do; if first.month then ofs=1; else ofs+1; /* other family size */ flag + (IndividualID not in IDS) /* check for new member */ + (ffs NE ofs)*last.month /* size change (e.g., from 3 to 2) */ ; end; end; flag = flag&1; output; keep GroupID flag; run; ... View more

Hi Paige.   To be 100% honest, I'm not sure how it worked.  The first track that it (font=) could help was this part of the documentation:   So I've started with simple: title font='Arial' height=11pt "Mischungsverhältnis Erbse : Gerste (%)"; which didn't work:   Then using good old "by try&error" approach I added: title font='Arial' "ABC" height=11pt "Mischungsverhältnis Erbse : Gerste (%)"; Which gave:   And then by replacing ABC with space the final version popped up.   But... I did some more experiments after you asked about explanation and here what I'v got: When I add "/bold" to the font name it "does not work" again: title font='Arial/bold' " " height=11pt "Mischungsverhältnis Erbse : Gerste (%)"; Result:   What's even more interesting adding, adding "/italic" (only) makes it BOTH italic and bold: title font='Arial/italic' " " height=11pt "Mischungsverhältnis Erbse : Gerste (%)";   But when you make it like this: title font='Arial/italic' " " height=11pt font='Arial' "Mischungsverhältnis Erbse : Gerste (%)"; works again (!)...     Another "interesting" behaviour, when I play with multiple different fonts, e.g., title font='Arial' " " font='Times New Roman' "Mischungsverhältnis Erbse : Gerste (%)"; the "Gesamt-Pflanzen-TM_25-4-24_NStufe.html" file has Arial in it, but the "default" SAS html document generated in WORK's sashtml1.htm file is Times... (even if there is no Arial at all... title font='Times New Roman' " " height=11pt "Mischungsverhältnis Erbse : Gerste (%)"; it displays Arial)   All the best Bart           ... View more

1) you have to have R installed on the server, 2) you have to have RLANG option in server's config, 3) you have to have R_HOME system variable set,  4) you have to have SAS/IML license.   In 3) you can set the the variable yourself in the SAS session by setting options: options set=R_HOME="/path/to/your/R"; eg: options set=R_HOME="/X/R/R-4.3.1/"; that also means you can have R installed locally in your home directory by yourself.   Bart ... View more

Looks like a school homework task to me.   data v_sort; input v1$ v2$ V3$ ; cards; A B C F K J D E F C K J K L W S V R ; run; proc contents data=v_sort out=list(keep=NAME varnum) noprint; proc sort data=list; by varnum; run; data _null_; set list end=_E_; call execute(cats('proc sort data=v_sort(keep=',name,') out=___',varnum,'; by _all_; run;')); if _E_ then call execute(cats('data v_sort2; merge ___1-___',varnum,'; run;')); run;   ... View more

TRUNCOVER  - thanks Tom, will use it!   Bart ... View more

I did it last year. I've submitted my SAS packages and SAS Packages Framework idea just to learn that since I'm not SAS customer but SAS user I'm not eligible...   Bart ... View more

And if you really have DSD-data then use colon operator(:)   Like this: data have2; infile datalines dlm="," dsd; input custid date :date9. ind_change balance XXX :$5.; /* XXX is really DSD-data */ format date date9.; cards; 111,01JAN2025,0,50,"A,B" ; Run; proc print; run; B ... View more

1) add MISSOVER 2) remove DSD 3) read the documentation for INFILE statement:  https://documentation.sas.com/doc/en/pgmsascdc/9.4_3.5/lestmtsref/n1rill4udj0tfun1fvce3j401plo.htm   data have1; infile cards MISSOVER; input custid date date9. ind_change balance; format date date9.; cards; 111 01JAN2025 0 111 02JAN2025 0 111 03JAN2025 0 111 04JAN2025 1 111 05JAN2025 0 111 06JAN2025 0 111 07JAN2025 0 111 08JAN2025 0 111 09JAN2025 1 111 10JAN2025 0 ; Run; data have2; infile datalines dlm=","; input custid date date9. ind_change balance; format date date9.; cards; 111,01JAN2025,0,50 111,02JAN2025,0,80 111,03JAN2025,0,10 111,04JAN2025,1,15 111,05JAN2025,0,30 111,06JAN2025,0,80 111,07JAN2025,0,90 111,08JAN2025,0,140 111,09JAN2025,1,130 111,10JAN2025,0,125 ; Run; ... View more

This thread looks like a copy on https://communities.sas.com/t5/SAS-Viya/Error-The-SAS-system-stopped-processing-this-step-because-of/td-p/955004   Consider deleting the duplicate.     ... View more