data have; input (a b c) (:$10.); cards; nrename test1.nm7 0test1.nm7 nrename test2.nm7 1test2.nm7 nrename test3.nm7 0test3.nm7 ; data want; set have; array t b--c; do over t; if first(t) in (' ', '1') then f=1; end; if f; drop f; run; array t b--c; This means we are grouping variables starting from b and anything between b and c, including c. HTH  ... View more

You can do it with SAS means: data files (keep=fname); rc = filename("mydir","/folders/myfolders"); did = dopen("mydir"); if did > 0 then do i = 1 to dnum(did); fname = dread(did,i); output; end; rc = dclose(did); run; Further SAS means (fopen, foptnum, foptname, finfo) can be used to retrieve details of files. ... View more

So you want to generate XX calls to NONMEM.  You want store them into more than one batch files. You could either specify the number of files you want and then calculate how many calls to write to each file. Or specify the maximum number of calls to write to each file. Let's do the latter.   %let num_runs=300; %let max_runs_per_file=100; data _null_; length fname $256 testname $50 ; do i=1 to &num_runs; testname = cats('TEST',i,'.CTL'); modi = mod(i,&max_runs_per_file); if 1=modi then batch+1; fname = cats('/path/','boot',batch,'.bat'); file bat filevar=fname ; if 1=mod1 then put 'REM ' batch= 'Starts with ' testname= ; put 'CALL NMGO ' testname '-MAXLIM=3' ; end; run; So this will create three files named boot1.bat, boot2,bat and boot3.bat in the directory /path/. Each file will have a first line that is a comment that shows which file it is and which test/run number it starts with followed by up to 100 calls to NONMEM. ... View more

Not sure what SAS product you are using. See if you can open the file through SAS using an open dialog box, or by navigating to it through a server path.   If you do manage to locate the file, then try getting properties on the file (e.g., right-click on it and then select properties), this would typically bring up the path to the file and then you could try copying and pasting it into your code to prevent any transcription errors.   If you are still having problems then consider if the file was created by some other process on your old machine and consider if that process needs to be run using the new machine.   If you can't navigate to the file (i.e., you can't find it as SAS has said) then I would suggest you need to create it.   Make sure any earlier processes that were supposed to have created it have completed correctly bu checking the logs or even just run them again if necessary.     Kind regards, Amir. ... View more

Well you create the final table each time from the most recent run so that is correct. Append the results to a table at the end if you want to keep it instead of using PROC SQL. proc append base=final data=alla_klmean&k; run; Note that you don't need the k in the data set names from what you've shown, it doesn't help anything if you're just stacking your tables. ... View more

You cannot call a macro that you haven't defined. You either need to include the macro definition (or have it in your autocall library).   But you don't need that macro. Replace %RecsInDS(&dsn,no_obs); with data _null_; call symputx('no_obs',no_obs); stop; set &dsn nobs=no_obs; run; ... View more

Since you do not have a variable subn1 (or subx1) in your input dataset, the array consists entirely of missing values. ... View more