Hi. See page 8-10 of this document http://www.ats.ucla.edu/stat/sas/library/ts274.pdf maybe it will clear up your question. I think asking a question such as "do I select anyone >0.2" is not something anyone can help. Anca. ... View more

Hi. This is a poorly written code (since I find it a bit confusing at how you want to assign data if a certain value is missing): *create data; data have; input FSA     $ Field1     $ Field2; cards; M5         A           45 M5         B           45 M5        A            36 M7A       A           98 M7         A           76 ; data test;     set have;     by fsa;     output;     flag = (fsa = "M5");     if fsa ne "M5V" & flag = 1 then do;         temp_fsa = "M5V";temp_field1 = field1;temp_field2 = field2;     end;     output; run; data want;     set test;     if temp_fsa ne "" then do;         fsa = temp_fsa;         field1 = temp_field1;         field2 = temp_field2;     end;     drop temp: flag; run; proc sort data = want nodupkey;by fsa field1 field2;run; This could be further improved, and possibly a macro could be used to allow for various FSA values... ... View more

Hmm, ok. alors je ne sais pas comment vous pouvez ajouter un pourcentage cumulé. Lorsque vous effectuez une freq proc sur la variable, vous ne recevez pas le pourcentage cumulé? Je peux être très confus. Je ne parle pas français, juste en utilisant Google Translate. ... View more

Salut. voici un lien (en anglais). Voir où il est dit: 33526 - Relative time calculations can be created in SAS® Web Report Studio reports that include a Time dimension Pour créer un calcul de temps relatif, procédez comme suit dans la boîte de dialogue Select Data:   1. Click the Custom tab. (  Cliquez sur l'onglet Personnaliser.) 2. Select a measure in the Available measures field.(Sélectionnez une mesure dans le domaine des mesures disponibles.) 3. Select one of the options in the Functions drop-down list. ( Sélectionnez l'une des options de la liste déroulante Fonctions.)   Vous avez la possibilité de sélectionner «cumulatif». J'espère que cela aide! ... View more

Here is some reference to help you out: SAS/GRAPH(R) 9.2: Statistical Graphics Procedures Guide, Second Edition You could use something like display = noticks; Anca. ... View more

Something like this could work. But you have to be careful with you first.keyword to reset all the lag values to . so it doesn't carry over the value from prior record... but this piece of code should get you started: proc sql;     select cat("prev_",name,"= lag(",name,");") into: my_list separated by " "     from dictionary.columns     where libname = "SASHELP" & memname = "CLASS" ; quit; data temp;     set sashelp.class;     &my_list.; run; All you need to change is the name of your library (WORK?) and the name of the dataset. Make sure you use CAPS. Good luck! Anca ... View more

Hi. Maybe this could work: data want;     set have;     new_string = translate(trim(prxchange("s/.*Commissions //", -1, string)), "_", ""); run; prxchange : will replace/change everything up to (including) the word Commissions with a blank space; trim : will remove the trailing blanks (so you don't attach an underscore at the end of the string; translate : will replace any blanks with underscore. Good luck! Anca. ... View more

Hi. I think the issues is that in the first instance - code below, you are not CALCULATING anything, you are just saving c as c2....but if you were to use  sum(c) as c2, or whatever other function...then the newly created c2 variable would actually hold a calculated value.Do you know what I mean? proc sql; create table F2 as   select     nom,     (c) as c2,      case       when calculated c2 > 0 then x**2       else 0     end       as x2   from F1; quit; I hope this helps. ... View more

Hi. How about this: data want;     set your_data;   retain temp;   if accnt_num ne . then temp = accnt_num;       else if accnt_num =. then accnt_num = temp; run; Good luck! Anca. ... View more