11-02-2016 03:46 PM
I have following code running in SAS,
proc adaptivereg data=imported_excel
SEED = 123
PLOTS = ALL
partition fraction(validate=0.25 test=0.25);
model Y = A B C;
is there anyway I can output the partition dataset, test,validate, respectively? Thanks.
11-02-2016 05:00 PM
Ok...I tried to run it to check, but it keeps crashing in SAS Studio.
If you generate an OUTPUT data set, does it contain a _ROLE_ variable that denotes what each observation was categorized?
You can generate the OUTPUT data set using something like
If you have partitioned the input data by using a PARTITION statement, then a character variable _ROLE_ is included in the output data set. For each observation the value of _ROLE_ is as follows:
11-04-2016 05:53 PM
proc adaptivereg data=imported_excel SEED = 123 PLOTS = ALL details=bases; partition fraction(validate=0.25 test=0.25); model Y = A B C; output out=dataset_role; RUN;
If this doesn't work post your log. I ran it using the example in the documentation and it worked fine.
11-08-2016 02:57 PM
Thanks for your reply. It does work, but I still got one question. So I am using 25% of the whole dataset as "test", the number of observations in my whole data set is 816, literally I should get 204 "test" observations. However, I found there were 211 observations pulled out as "test" with the code above, is there anything wrong with this? Not a big deal, just curious, thanks!