BookmarkSubscribeRSS Feed
Saifulinfs
Fluorite | Level 6

 I am running Proc Surveylogistic where I have created spline variables through the effect statement. However, unlike proc logistic, this procedure (proc surveylogistic)  doesn't support the outdesign= option. Could you help me how to get access / retrieve the spline design variables constructed through the effect statement?

3 REPLIES 3
sbxkoenk
SAS Super FREQ

Hello,

 

I believe you can do it in two steps.

(Not sure as I am not aware of peculiarities in survey-procedures)

 

If you use the OUTDESIGN= option in PROC GLMSELECT, you would obtain the spline basis. PROC LOGISTIC and PROC GLIMMIX have the similar functionality (outdesign=). They all support the EFFECT statement.

 

Then use the design variables (as effects --> right-hand side x-variables) directly in the MODEL stmt of PROC SURVEYLOGISTIC. (you don't need EFFECT stmt anymore).

 

Tip : it can be a bit of a challenge to specify the model correctly. Keep going until you get exactly the same output without EFFECT statement (2 steps, but 1st step having EFFECT stmt of course) than with EFFECT statement (all in one step).

 

Good luck,

Koen

Saifulinfs
Fluorite | Level 6

Thank you but if I use the spline variables directly in the model statement, How I am going to allow nonlinear effect? Including the spline variables directly as right hand side variables in the model statement would assume linearity with the logit of the outcome

SAS Innovate 2025: Call for Content

Are you ready for the spotlight? We're accepting content ideas for SAS Innovate 2025 to be held May 6-9 in Orlando, FL. The call is open until September 25. Read more here about why you should contribute and what is in it for you!

Submit your idea!

What is ANOVA?

ANOVA, or Analysis Of Variance, is used to compare the averages or means of two or more populations to better understand how they differ. Watch this tutorial for more.

Find more tutorials on the SAS Users YouTube channel.

Discussion stats
  • 3 replies
  • 509 views
  • 0 likes
  • 3 in conversation