Im just trying to understand the effect on the beta estimates of multicollinearity in a regression. I understand that it effects the variances of the beta estimates and thus their tests for significance but am still unclear about the biasedness and consistency of the betas as in some places I read that theyre unbiased and others like here it seems to suggest that they are. I assume that they're not consistent as the variance is inflated here but am not positive. I wasnt able to find an answer to this on the site but if there is please provide the link.

In the link above, I am referring to this specific sentence: "More importantly, the usual use of regression is to take coefficients from the model and then apply them to other data. Since multicollinearity causes imprecise estimates of coefficient values, the resulting out-of-sample predictions will also be imprecise."

  • 1
    $\begingroup$ "Imprecision" and "biasedness" are completely different characteristics. The theorems about OLS estimates being unbiased assume nothing about whether the explanatory variables are orthogonal or not. $\endgroup$ – whuber Nov 14 '19 at 21:54

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Browse other questions tagged or ask your own question.