PM,   I really appreciate the code, and it works great.  However, the techniques seem to be very, very computationally demanding (as you alluded to when you said "Maybe there are better/faster ways").  Do you have any recommendations to speed up the process? Or should I post another question?   Thanks ... View more

The multiple asterisks may have messed things up if you missed a quote or comma somewhere. ... View more

Then you need a macro with nested %DO loops for the multiple variables. Since you have not provided any more details, there's really not much detail I can provide back to you. ... View more

1: Post code on this site in code boxes. The main message windows reformat text, sometimes making things unreadable.   Any process involving 1,000s of data sets will take some time. Another approach is to write thousands of proc append statements. This could be added as the last bit replacing call execute ('data perm.tryingloop ; set '||dn||' perm.tryingloop ; run;');         with call execute ('proc append base=perm.tryingloop data=' ||dn||';run;');   But I doubt there will be much improvement in time if any as the sheer amount to IO start and stop is likely to more time intensive that the actual combination of the data.   Another option might be to FTP or download the entire contents of the source URLS to a common local folder and use a data step to read all the txt files at one time.   Filename myfile "c:\path\*.txt"; for example on a Windows system will read ALL of the files with TXT as the extension in a given folder when used in an Infile statement. There are options to keep the source filename (Filename= ) on the infile if you need to know the source. That could be used for by group processing. Note that the Filename= variable is temporary so would need to assign the value to a data step variable with sufficient length to hold the path and file name for use as a by variable.       ... View more