Detect multicollinearity between predictors using the variance inflation factor.
A VIF of 10 corresponds to an R-squared of 0.9 between one predictor and the rest, and inflates that coefficient’s standard error more than threefold. Coefficients become unstable long before predictions do. Multicollinearity does not harm predictive accuracy but makes individual coefficients uninterpretable, which matters whenever the goal is explanation rather than prediction.
Variance Inflation Factor
VIF = 1 ÷ (1 − R²ⱼ); standard errors inflate by √VIF
VIF = 1 ÷ (1 − R²ⱼ); standard errors inflate by √VIF A VIF of 10 corresponds to an R-squared of 0.9 between one predictor and the rest, and inflates that coefficient’s standard error more than threefold. Coefficients become unstable long before predictions do.
Multicollinearity does not harm predictive accuracy but makes individual coefficients uninterpretable, which matters whenever the goal is explanation rather than prediction.
This calculator takes 3 inputs: R-squared regressing this predictor on the others, Number of predictors, Standard error without collinearity. The pre-filled defaults are a realistic starting point — replace them with figures from your own environment for a result you can act on.