cancel
Showing results for 
Show  only  | Search instead for 
Did you mean: 
Try the Materials Informatics Toolkit, which is designed to easily handle SMILES data. This and other helpful add-ins are available in the JMP® Marketplace
Choose Language Hide Translation Bar
jmpuser1234
Level I

How to check for multicollinearity in Ordinal logistic regression

I have an ordinal logistic regression model with 12 continuous independent variables. How can I check for multicollinearity of the independent variables (I can find anything on VIF’s for ordinal logistic models)?

1 ACCEPTED SOLUTION

Accepted Solutions
jiancao
Staff

Re: How to check for multicollinearity in Ordinal logistic regression

VIF uses the independent variables only, not the dependent variable. So you can just set your DV from ordinal to continuous and run the standard least squares fit to get the VIFs.

View solution in original post

2 REPLIES 2
jiancao
Staff

Re: How to check for multicollinearity in Ordinal logistic regression

VIF uses the independent variables only, not the dependent variable. So you can just set your DV from ordinal to continuous and run the standard least squares fit to get the VIFs.

jmpuser1234
Level I

Re: How to check for multicollinearity in Ordinal logistic regression

Thanks for your reply. I thought so too after looking at the formula for computing VIF's...but I ran two models with two different dependent variables and got different (but similar) VIF's...so I thought I must be missing something...so I went back to these models again and realized the second dependent variable had about 200 missing values so that explains the different VIF's.