Odds ratios in Proc Logselect

andreas_zaras — Mon, 28 Nov 2022 10:47:06 GMT

Hello,

I run a logistic regression using the interace of SAS Studio. I get the code at the ned of the question. How can i i add odds ratios in the output?

I add the oddratio statement as described in the following link but i get the error:

91 link=logit clb;

92 oddsratio;

---------

180

ERROR 180-322: Statement is not valid or it is used out of proper order.

https://documentation.sas.com/doc/en/pgmsascdc/v_023/casstat/casstat_logselect_examples08.htm

Thanks in advance,

Andreas

ods noproctitle;

proc logselect data=CASUSER.PVA97NK_26_VIYA_PARTITION;
	partition fraction(validate=0.3);
	class DemCluster DemGender DemHomeOwner StatusCat96NK;
	model TARGET_B(event='1')=DemCluster DemGender DemHomeOwner StatusCat96NK 
		DemAge DemMedHomeValue DemMedIncome DemPctVeterans GiftAvg36 GiftAvgAll 
		GiftAvgCard36 GiftAvgLast GiftCnt36 GiftCntAll GiftCntCard36 GiftCntCardAll 
		GiftTimeFirst GiftTimeLast 'Partition Indicator'n PromCnt12 PromCnt36 
		PromCntAll PromCntCard12 PromCntCard36 PromCntCardAll StatusCatStarAll / 
		link=logit clb;
	selection method=stepwise
     (stop=sbc choose=sbc) hierarchy=none;
	output out=CASUSER.sot predicted;
run;

ods noproctitle;

proc logselect data=CASUSER.PVA97NK_26_VIYA_PARTITION;
	partition fraction(validate=0.3);
	class DemCluster DemGender DemHomeOwner StatusCat96NK;
	model TARGET_B(event='1')=DemCluster DemGender DemHomeOwner StatusCat96NK 
		DemAge DemMedHomeValue DemMedIncome DemPctVeterans GiftAvg36 GiftAvgAll 
		GiftAvgCard36 GiftAvgLast GiftCnt36 GiftCntAll GiftCntCard36 GiftCntCardAll 
		GiftTimeFirst GiftTimeLast 'Partition Indicator'n PromCnt12 PromCnt36 
		PromCntAll PromCntCard12 PromCntCard36 PromCntCardAll StatusCatStarAll / 
		link=logit clb;
	selection method=stepwise
     (stop=sbc choose=sbc) hierarchy=none;
	output out=CASUSER.sot predicted;
run;

ods noproctitle;

proc logselect data=CASUSER.PVA97NK_26_VIYA_PARTITION;
	partition fraction(validate=0.3);
	class DemCluster DemGender DemHomeOwner StatusCat96NK;
	model TARGET_B(event='1')=DemCluster DemGender DemHomeOwner StatusCat96NK 
		DemAge DemMedHomeValue DemMedIncome DemPctVeterans GiftAvg36 GiftAvgAll 
		GiftAvgCard36 GiftAvgLast GiftCnt36 GiftCntAll GiftCntCard36 GiftCntCardAll 
		GiftTimeFirst GiftTimeLast 'Partition Indicator'n PromCnt12 PromCnt36 
		PromCntAll PromCntCard12 PromCntCard36 PromCntCardAll StatusCatStarAll / 
		link=logit clb;
	selection method=stepwise
     (stop=sbc choose=sbc) hierarchy=none;
	output out=CASUSER.sot predicted;
run;

Re: Odds ratios in Proc Logselect

PaigeMiller — Mon, 28 Nov 2022 14:08:24 GMT

The documentation at your link says:

This documentation is for a version of the software that is out of support. Select a version from the version selector in the banner, or access the latest documentation.

At this link: https://documentation.sas.com/doc/en/pgmsascdc/9.4_3.4/casstat/casstat_logselect_toc.htm there is no ODDSRATIO statement.

Re: Odds ratios in Proc Logselect

StatDave — Mon, 28 Nov 2022 15:06:57 GMT

An ODDSRATIO statement is available in PROC LOGSELECT in recent versions of SAS Viya.

Re: Odds ratios in Proc Logselect

andreas_zaras — Mon, 28 Nov 2022 18:26:33 GMT

What about sas viya for learners?

Re: Odds ratios in Proc Logselect

StatDave — Tue, 29 Nov 2022 16:02:24 GMT

It should work there since I believe it will give you access to the current or recent release of SAS Viya.

topic Re: Odds ratios in Proc Logselect in Statistical Procedures

Odds ratios in Proc Logselect

Re: Odds ratios in Proc Logselect

Re: Odds ratios in Proc Logselect

Re: Odds ratios in Proc Logselect

Re: Odds ratios in Proc Logselect