Linear Modelinghard
0:00.0

In a multiple regression model Y=Xβ+ϵY = X\beta + \epsilon, what happens to the variance of the OLS estimator β^\hat{\beta} if the predictors become perfectly collinear?