Re: SAS PROC LOGISTIC TESTS

AgEconomist wrote:
On Jan 6, 1:38 pm, Ryan <Ryan.Andrew.Bl...@xxxxxxxxx> wrote:
On Jan 6, 12:25 pm, AgEconomist <matttbog...@xxxxxxxxx> wrote:

Are there any commands in SAS that would test a logit model in PROC
LOGISTIC for multicollinearity, heteroskedasticity, or serial
correlation ? PROC REG has the VIF, DW options in the model statement
but not in PROC LOGISTIC. I could probably write a routine, but
frankly, I’m not even sure about how to get the ‘residuals’ necessary
for some of these tests. I know that RESDEV and RESCHI in the model
statement gives residuals, but can I treat these the way I would treat
residuals from OLS?
I do not know of a way to test for multicollinearity in PROC LOGISTIC.
Since multicollinearity is a property of the IVs, you can just use
PROC REG. -Ryan

Thanks- that never occurred to me. I also posted to another group, and
that was the common advice given there too.

Multicollinearity is a product of the IVs plus the DV, so PROC REQ will just be an approximation. Most of the information is given by the IVs but not all.

There are several tools in R to help, related to redudancy analysis and variable clustering, which also don't use the DV.

Frank
.

Relevant Pages

  • Re: SAS PROC LOGISTIC TESTS
    ... PROC REG has the VIF, DW options in the model statement ... statement gives residuals, but can I treat these the way I would treat ... I do not know of a way to test for multicollinearity in PROC LOGISTIC. ...
    (sci.stat.consult)
  • Re: SAS PROC LOGISTIC TESTS
    ... PROC REG has the VIF, DW options in the model statement ... statement gives residuals, but can I treat these the way I would treat ... I do not know of a way to test for multicollinearity in PROC LOGISTIC. ...
    (sci.stat.consult)
  • Re: multiple regression assumptions SAS
    ... To test the assumption in multiple regression is that the errors are ... you would output the residuals from PROC REG ...
    (sci.stat.math)
  • SAS PROC LOGISTIC TESTS
    ... Are there any commands in SAS that would test a logit model in PROC ... PROC REG has the VIF, DW options in the model statement ... statement gives residuals, but can I treat these the way I would treat ...
    (sci.stat.consult)
Loading