For controlling annotate values for what you're doing I suggest investigating the annotate macros %ANNOMAC and %SLICE. A bit more code but may provide more what you are looking for. ... View more

Your ARR contains a single string. Before dealing with options on handling, it does help to how the how thing is used. For most purposes in a macro both the quotes around your values and the commas are likely to be unneeded. ... View more

I'd suggest selecting for the non-nulls and nulls separately and using a UNION with the nulls in the second part of the clause. ... View more

One guesses that the syntax you attempted maybe should have read: IF CROP = 'corn' THEN DELETE; /* must match the spelling and case of the value */ unless you intended to check if the value of the variable CROP was the same as the variable CORN. Unless you really want to proliferate data sets, specifying a WHERE statement such as: WHERE CROP='corn' or DATA=SOILS (where=(crop='corn')) in the proc call, will accomplish the subset for any procs that do not directly support a where statement ... View more

Your code is running forever because of this line: IF CODE = 500 THEN DO UNTIL (CODE=100); Since you have nothing changing the value of CODE the first time CODE=500 it goes into the loop and never exits because CODE is always 500. ... View more

The solution may be to use SAS/AF in a server mode. I haven't used AF for awhile but one of the internet options was to have a SAS server application do all of the heavy work and the results were presented as HTML or similar output. The users didn't need to have any SAS install, just a web browser. An example of a site run by the US Census is: http://www.census.gov/hhes/www/cpstc/cps_table_creator.html This site lets you build requests for categories of data from the Current Population Survey and even set a few customizations. The parts I vaguely remember on this topic had several different ways to approach the Web/SAS interface. ... View more

If you want all of the output to align on the left margin you can also use OPTION NOCENTER: ... View more

If these are existing SAS data sets and they do not have the same name you can use a LIBNAME statement that references all of the directories. in windows it might look like: libname mylib ('..../2010/08' '..../2010/09' '..../2010/10' '..../2010/11'); The quotes may not be needed if you have no spaces in the paths, but I use them habitually just in case. The the set statment might look like: set mylib.set1 mylib.set2 mylib.set3 mylib.set4 ; ... View more

Does the KEYS window show an entry for CTL END in your OS? KEYS settings may be involved with the "classic" program window. ... View more

I'm not sure it this will completely suit your purpose but here is one approach. A small example data set. data junk; infile datalines truncover; input v1-v5 ; label v1 = 'Label 1' v2 = 'Label 2' v3 = 'Label 3' v4 = 'Label 4' v5 = 'Label 5' ; datalines; 1 2 3 4 5 6 7 8 9 10 11 12 13 ; run; Manually transpose the data creating a page variable and using the labels of the individual variables to conditionally create a categorical variable for non-missing values. data junk2 (keep=page label value); set junk; array v v1-v5; do i= 1 to dim (v); page= _n_; if v ne . then do; label= vlabel(v); value= v; output; end; end; run; This will provide a single class or group variable, LABEL in this case with the values in an analysis varible and PAGE something to provide a break on changing values. ... View more

The difficulty may be with Excel. Excel will remap colors to the "closest" one in its current default color pallet. Your best bet may be to pick a style that has fairly different close to default Excel colors for the various components so they are less likely to be drastically remapped. ... View more

Put the LENGTH statement AFTER the SET but before where the variable is used. The new variables will be at the "end" of the data. ... View more

I'm not sure I understand completely what you are attempting but it may be that PROC TABULATE with the ALL predicate, which gives summaries of items in a class, may be a place to look. ... View more

It has been awhile since I used SPSS but the option to export to SAS had a check box to export the formats and labels. This generated separate SAS code files with PROC FORMAT and LABEL statement syntax. If you didn't check the box no joy for labels or fromats. ... View more

Probably not needed as a response to the original post at this time but the FORMAT associated with the axis variable also would be a place to look to increase readability especially with dates. The example looks like a MMDDYY10. or MMDDYY8. would have fit and been readable. Depending on the number of values they may have then even appeared horizontally. ... View more