I'm receiving a SAS dataset with five integer variables (x1-x5) with different custom formats already applied. I also have the formats catalog. I want to create character variables x1_DEC through x5_DEC that contain just the decodes. I would normally solve this by using x1_DEC=put(x1, formatA.); x2_DEC=put(x2, formatB.);   but that requires hardcoding the format names into the program. Is there a way to do this using vformat or some such function so that I can write something like x1_DEC=put(vformat(x1)); that would dynamically read the currently assigned format? ... View more

I'm trying to write code to modify the existing labels of a dataset by adding a suffix. This code is working for me: %let ds=DS; %let Suffix=MySuffix; proc sql; CREATE TABLE newlabels AS SELECT "LABEL "||trim(name) || "='" || trim(label) || "&Suffix';" as labelcode FROM dictionary.columns WHERE libname='WORK' and memname="&ds" ORDER BY varnum; quit; data _null_; file "H:\sas\Relabel.txt" notitle; set newlabels end=lastobs; if _n_= 1 then do;   put "DATA &ds;";   put "  SET &ds;";   end; put labelcode; if lastobs=1 then do;   put "run;";   end; run; %include "H:\sas\Relabel.txt"; , until I reach a label that contains an apostrophe. That creates a big unmatched quote mess. Is there a function that I can place around 'trim(label)' in line 5 that will double any internal apostrophes? Thanks. ... View more

I have a data set that has person, day, and events. The data is sparse; there are no observations showing events=0. I want to fill it out with all of the days from 01jan2012 to 31mar2012 for each person with events=0 so that I can eventually use proc summary or a data step to create monthly and global sums by person, where I'll need to report months with zero events. data have; input person day : date9. events; format day date9.; datalines; 1 03jan2012 1 1 15jan2012 2 1 01mar2012 1 2 02feb2012 3 2 01mar2012 1 ; run; want: person day events 1 01jan2012 0 1 02jan2012 0 1 03jan2012 1 1 04jan2012 0 ... 1 31mar2012 0 2 01jan2012 0 ... 10 31mar2012 0 How can I use a data step to loop through all people (1 to 10) and from 01jan2012 to 31mar2012 and append the missing observations with events=0 ? Thanks, pteranodon ... View more