Variance Inflation Factor
- Subscribe to RSS Feed
- Mark Topic as New
- Mark Topic as Read
- Float this Topic for Current User
- Bookmark
- Subscribe
- Mute
- Printer Friendly Page
- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Permalink
- Notify Moderator
In Linear regression (multiple), I am using predictor variables that I know are corelated. Now to understand their effect on the model I wanted to look at VIF. Please guide how to get VIF in Alteryx and also if there are other/better ways to assess impact of multicolinearity on the model.
Solved! Go to Solution.
- Labels:
- Predictive Analysis
- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Permalink
- Notify Moderator
Hi Ashish,
Attached is a yxzp file that contains an example module where a model is run and the (G)VIFs are calculated and appended onto the Report that comes out of (in this case) a logistic regression tool. For continuous variables a traditional VIF (variance inflation factors) is reported, for categorical variables the Generialized Variable Inflation Factors of Fox and Monette are reported. Once you extract the workflow, the workhorse macro (vif.yxmc) will be in a folder called vif_example.
Dan
- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Permalink
- Notify Moderator
A quick follow on, the vif.yxmc macro will work wfor models created with the Linear, Logistic, Count, and Gamma Regression tools.
- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Permalink
- Notify Moderator
Thanks a lot DrDan! Really helpful!.
