The variable planname can also use the IDlabel and PROC EXPORT will print the label not the variable name.  @ballardw wrote: The output data set probably has the group variables correctly. The issue is that the ACROSS variable  may no longer be there.   As I mentioned, you have not provided any example of what the actual output should look like.   You may be able to get something like what you want with Proc Summary (or means) and Transpose IF the value of planname can be a valid SAS variable name. And example using data you should have available: proc summary data= sashelp.class nway; class age sex name; var weight; output out=work.sum sum=; run; proc transpose data=work.sum out=work.trans; by age sex; id name; var weight; run; The _name_ variable could be dropped from work.trans.   The CLASS variables in proc summary would be all of the group variables and the last element would be the Across variable. This won't quite work if there are more than one variable to summarize as the "across" variable can only be used for one column name out of proc transpose.   ... View more

Hi @TimH   Right now there is no SAS/ACCESS Interface to JDBC. You may be able to get PROC GROOVY to work. Take a look at this SGF paper from Lilyanne Zhang. Accessing DBMS with the GROOVY Procedure and a JDBC Connection   Best wishes, Jeff   ... View more

I think this is in the wrong area, but I can't find where to change it. ... View more

Well when looking at the data, the shortest duration is  .01 seconds (one-one hundredth) and the longest is 2748.97 seconds. This is during an overall time period of 50400 seconds (an overnight period, from 5 PM to 7 AM).  I've eliminated the idea of this chart (at least as a wall chart) with the following logic (perhaps) eliminate anything with duration under 1 second. One second has to equal approximately 10 characters (8-character name plus some space on each side). So, one second equals 10*n  where n = the width of a character. the entire time period then equals 50400*10*n = 504000*n ;   n would have to be unreadably small even if the chart were to go on a 12 foot wall.  For 504000 to fit in 12 feet means it fits in 144 inches,  therefore n = 144/504000 = .00028  inches. We'd need a microscope! I'll come up with some sort of scrolling idea, using HTML and a projector, I guess.  We can close this topic. ... View more

In the Communities main page, look in the Actions box in the bottom right: https://communities.sas.com/community/support-communities ... View more

I didn't realize that would create a link. Uploading it this time. ... View more

Thanks, your last example works ... View more

Hi:   You would need to change the syntax of your COLUMN statement. The PROC REPORT documentation shows how to use the comma operator with parentheses to group items under an ACROSS usage variable. Showing the code you've tried and a sample of your data would be very helpful. In addition, knowing your final ODS destination of choice (part of your code) would be helpful too. cynthia ... View more

The procedure only supports one data set per graph.  So, you have to merge the data.  It is OK to have two columns (X & Y) for the upper graph and two different columns (X2, Y2) for the lower.  As long as X and X2 are of the same type, the axes can be merged by the graph.  We do that with the survival plot.  The survival graph is with x=Time and the lower at risk table graph is using x=TAtTisk. ... View more