0

It is known that severe multicollinearity is a problem because it can increase the variance of the coefficient estimates and make the estimates very sensitive to minor changes in the model. The result is that the coefficient estimates are unstable and difficult to interpret.

What are the mathematical reasons for that phenomenon?

Amit S
  • 27
  • 7
  • Try this previous answer for some of the intuition: https://stats.stackexchange.com/questions/1149/is-there-an-intuitive-explanation-why-multicollinearity-is-a-problem-in-linear-r – Aaron Springer May 21 '21 at 20:31
  • See especially [this answer in the duplicate thread](https://stats.stackexchange.com/questions/1149/is-there-an-intuitive-explanation-why-multicollinearity-is-a-problem-in-linear-r/34721#34721) – kjetil b halvorsen May 22 '21 at 15:35

0 Answers0