Turn on suggestions

Auto-suggest helps you quickly narrow down your search results by suggesting possible matches as you type.

Showing results for

- JMP User Community
- :
- Discussions
- :
- Re: How to find VIFs in a Nominal Logistic Regression? Using JMP Pro 14

- Subscribe to RSS Feed
- Mark Topic as New
- Mark Topic as Read
- Float this Topic for Current User
- Bookmark
- Subscribe
- Printer Friendly Page

Highlighted
##

- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Get Direct Link
- Email to a Friend
- Report Inappropriate Content

How to find VIFs in a Nominal Logistic Regression? Using JMP Pro 14

Feb 12, 2020 9:43 AM
(144 views)

I've tried right-clicking in the table of parameter estimates and adding as a column, but the option is not there. I also tried Help -- Statistics Index -- multicollinearity -- launch.

Is there anything else I could try?

Thanks!

2 REPLIES 2

Highlighted
##

- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Get Direct Link
- Email to a Friend
- Report Inappropriate Content

Re: How to find VIFs in a Nominal Logistic Regression? Using JMP Pro 14

Hi @cwr41

have a look at this discussion.

How to examine VIF in a generalized regression report?

let us know if it helped.

ron

- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Get Direct Link
- Email to a Friend
- Report Inappropriate Content

Re: How to find VIFs in a Nominal Logistic Regression? Using JMP Pro 14

I do not find VIF as an available column in the Parameter Estimates table in GenReg either.

I tried to obtain the model matrix X without going to a lot of trouble but these platforms do not provide direct access to it.

Since the VIF is about the correlation among the estimates and it has nothing to do at all with the response Y, you might use Fit Least Squares to regress your linear predictor against a continuous Y in your data table. If you don't have one, you could simulate one. Any random response will do. You are not interested in the relationship (e.g., R square) between Y and X but just within the X matrix. Now you can select VIF in the Parameter Estimates.

Kind of round-about, I know. There might be a reason to intentionally omit VIF but I am not aware of it.

Learn it once, use it forever!