Multicollinearity & VIF

When X variables are highly correlated, the model cannot separate their individual effects — standard errors inflate and coefficients become unstable. VIF (Variance Inflation Factor) quantifies this for each predictor:

  • VIF < 5: clean — no multicollinearity concern
  • VIF 5–10: moderate — review variable selection
  • VIF > 10: severe — remove or combine the variable
Fix:
  • Remove one of the correlated variables
  • Combine them (ratio, difference, index)
  • Collect more diverse data
Try it in the app
Try: Run regression with 2+ X variables → check the VIF column in Results → Coefficients table. Values >5 deserve attention.
Watch this on our YouTube channel

Run this on your own data

Free tier, no signup needed. Works offline in the browser.

Open the app