Logistic Regression using CATMOD

[Proshi]

Dear all,

I would like to know if I could estimate Odds ratio with a Catmod procedure?

Thank you,

Darina Proshenska

[Ksharp]

I am not sure .

CONTRAST '1 vs. 4' A 2 1 1  /  EXP ;

[Proshi]

I will try this. Thank you for your help!

Darina

[Rick_SAS · Accepted solution]

Yes. Use the LOGLIN statement  to specify the model effects and put the _RESPONSE_ keyword on the right-hand side of the MODEL statement. The parameter estimates table contains the estimates on a log scale. You will need to combine them and exponentiate them to get the odds ratios. I wasn't able to find an online example, but there is an example in the "Loglinear Models" chapter of Stokes, Davis, and Koch, "Categorical Data Analysis" book.

[Proshi]

Okay. I'll check this.

Thanks, Rick,
Darina Proshenska

[Rick_SAS]

@StatDave gave you some valuable advice about using PROC LOGISTIC instead of PROC CATMOD. Please consider following his advice.

[Proshi]

Yes. Thank you!