How do I get ROC information from just the final step in PROC LOGISTIC with Stepwise Selection?

[oakHILLS68]

In PROC LOGISTIC (v.9.4) with SELECTION=STEPWISE, if you ask to output the ROC curve data (OUTROC), you get results for every step.  My question: Is there a way to restrict the output to just the final step?  Failing that, is there a way that I can use ODS to output the total number of steps so that I can subset the OUTROC dataset?  [I suppose I could use proc freq to find the maximum number of steps and then try to pass that maximum number to a DATA step.  But that's very kludgy.]

My current code:

proc logistic data=hem;
model outcome(ref='Alive') = IL_1ra IL_6 IL_8 IL_10 Eotaxin IP_10 MCP_1 / outroc=hem selection=stepwise;
output out=outhem predicted=predhem xbeta=xbhem;
run;

Thanks,

Dennis Hanseman

[Reeza]

If you're just interested in the final results, why not just use the final model and remove the selection option entirely?

[oakHILLS68]

That's a good idea, but my code is in a macro. I need to run it multiple times with different macro parameters.

Thanks.

[Reeza]

So pipe the selected variables from the previous logistic regression to the next step using macro variables.

[FreelanceReinh · Accepted solution]

---

@oakHILLS68 wrote:

(...) is there a way that I can use ODS to output the total number of steps so that I can subset the OUTROC dataset?  [I suppose I could use proc freq to find the maximum number of steps and then try to pass that maximum number to a DATA step.  But that's very kludgy.]

---

Hello @oakHILLS68,

You can do the selection in one DATA step;

data want(drop=_s);
if _n_=1 then set hem(rename=(_step_=_s)) point=n;
set hem nobs=n;
if _step_=_s;
run;

[oakHILLS68]

Yes! Thanks very much.