And if your data isn't very clean and the spelling may have varied case or have leading blanks. Proc format ; value $coupf (just upcase ) 'ANNUALl'      = '1' 'MONTHLY'     = '4' 'SEMI-ANNUAL' = '2' 'QUARTERLY'   = '3' other        = '2'; run; ... View more

If I know I'm going to be reading the same layout file multiple times I usually take the data step code generated by PROC IMPORT and modify it to reduce such suprises. That way I control any informats, lengths of string variables, variable names, labels and such. ... View more

I don't know if this applies to you but when I upgraded to SAS 9.2.3 I had an issue with a custom style not working and rebuilding based on a SAS supplied style did not help. I had to get to the source of the style to find out that the SAS supplied version had changed how it was addressing color and fonts in the base style and my existing code wasn't compatible. I had to to modify the code to align with with the new base definition method. ... View more

You might want to add a LRECL= 32767 parameter on your INFILE statement (probably doesn't have to be as big as 32767 but that provides room for lots of difference in columns). The lengths of the variables with attributes listed totals to 248 columns and the added numerics are likely pushing it over the default 256 columns. ... View more

My approach would be to ensure the dates are SAS date values and then use a custom format. Proc format; value MyYear '01OCT2010'd - '30SEP2011'd = '2010/2011' '01OCT2011'd - '30SEP2012'd = '2011/2012' ; run; Then in most summary procs assigning the format MYYEAR to the date will give you the grouping AND the desired display text. ... View more

For use with ORDER=DATA sometimes it helps to sort the data so the order is as expected. ... View more

It may be helpful to let us know how you are sending the data to Excel and whether the output works as intended when sent to the SAS results window. Ar you implying you had footnote statements greater than 10 such as: FOOTNOTE11 "more of my footnot text appears here"; There is a limit of 10 FOOTNOTE statements last time I checked. ... View more

You might try DBMS= DBFMEMO to read data with memo fields, which sounds like your data. ... View more

I can only find examples not documentation. This makes me wonder what other things may be buried in GTL that we can't find. ... View more

X is a key word. It is a command used to run operating system commands. ... View more

Your options might involve placing text from the MARKER font or use PROC GFONT to create a graphic font from scratch to show what you want . The %LABEL annotate macro would be the way to use either of these. ... View more

I am guessing that you are trying to create the macro variables parmSite and parmSubj from the values of siteid and subjid. You are getting the errors because siteid and subjid are NOT macro variables. In the data step you are looking for: call symput(parmsite,siteid); call symput(parmsubj,subjid); run; then your fpath assignment may work. Depending on length assignments and values you may want to use strip to remove unwanted blanks such as call symput(parmsite,strip(siteid)); if you get unwanted spaces in the resolved value for fpath. ... View more

I believe the behavior you are seeing is because of the EXCEL data engine not SAS. I know of two work arounds 1) don't let the Excel sheet get populated with the (.) 2)save each sheet as CSV and then the problem either goes away or the guessingrows will work. ... View more