Correlated regressors, what actually breaks?
Asked at Two Sigma, Citadel
You regress returns on two signals that are highly correlated with each other (say ).
What does multicollinearity do, and not do, to OLS? Derive the variance-inflation effect, list the symptoms, and give the remedies.
Show a hint
Write the sampling variance of one coefficient in terms of how much of that regressor is explained by the others.
Your answer
This one is open-ended. Work it through, then check your reasoning against the full solution.