01-26-2016 01:47 PM
I am using SAS University edition and I am having issues trying to compile all my type 3 test of fixed effects tables together in one file. I was given this code a while ago:
ODS OUTPUT SolutionF=FixedEffs Tests3=FTests
I believed this worked in SAS 9.4 on a different computer, but it is not working on my computer for SAS University edition. I also tried going to the work library and clicked on the FTESTS folder, but only one type 3 test of fixed effects table was shown out of my 10 analyses (it was the table from the first analysis). Any help would be greatly appreciated!
01-26-2016 03:16 PM
What does not working mean?
If you received an error please post it, as well as the code.
01-26-2016 03:54 PM
I think you need to provide a different name for each output set and include a permanent library if you want them to persist across sessions.
Something like this for the first and then increment the number for each run:
ODS OUTPUT SolutionF=Ftests.FixedEffs_Run01 Tests3=Ftests.FTests_Run01;
ODS OUTPUT SolutionF=Ftests.FixedEffs_Run02 Tests3=Ftests.FTests_Run02;
OR immediately after each analysis append the results to appropriate datasets.
01-26-2016 03:59 PM
Thank you for the responses Reeza and ballardw. I figured out the problem. First my data was in wide format, so I converted it to long format in excel. Does anybody know how to do that conversion in SAS, for future reference?
Next, the ods statement was incorrect. See my correction below
infile '/folders/myshortcuts/methylation/GPR126.csv' dsd firstobs=2;
input pig $ sex $ gest $ post $ sow $ sire $ gpr126 location $ position $ methperc;
proc print data=gpr126_methdiff;
by position gest;
proc mixed data=gpr126_methdiff;
class pig sow gest post sex sire;
model methperc = gest|post|sex; random sow(gest); random sire;
ods output Tests3=Tests3;
Thanks again for the help!
01-26-2016 04:07 PM
There are lots of threads in this forum about transposing data from wide to long and long to wide. Search for TRANSPOSE will find many of them.
If you run into specific problems, create a new thread with some example input data and what the result should look like. It really helps if you can provide data in the form of a data step or demostrate with one of the SAS supplied data sets that usually reside in the SASHELP library.