How to Deal With Multicollinearity (VIF above 10)

Описание: What to do about multicollinearity (i.e., a variance inflation factor above 10) in a multiple regression? This video will show you multiple options for handling multicollinerity:

0:00 Start
0:43 Ignoring it (based on the paper by O'brien)
3:29 Removing multicollinearity by removing predictors
3:58 Removing multicollinearity by pooling predictors
4:32 Adressing multicollinearity in moderation analysis and polynomial regression (structural multicollinearity)
5:15 Using penalized regression (ridge regression, lasso regression)
6:00 Avoiding specification errors

O'brien's paper:
O’brien, R. M. (2007). A caution regarding rules of thumb for variance inflation factors. Quality & Quantity, 41(5), 673-690.

Paper about mean centering:
Dalal, D. K., & Zickar, M. J. (2012). Some common myths about centering predictor variables in moderated multiple regression and polynomial regression. _Organizational Research Methods_, _15_(3), 339-362.

Papers about ridge regression:
Marquardt, D. W., & Snee, R. D. (1975). Ridge regression in practice. _The American Statistician_, _29_(1), 3-20.

Wilcox, R. R. (2019). Multicolinearity and ridge regression: results on type I errors, power and heteroscedasticity. _Journal of Applied Statistics_, _46_(5), 946-957.

Videos about ridge regression and lasso regression:
   • Regularization Part 1: Ridge (L2) Regression  
   • Regularization Part 2: Lasso (L1) Regression