0
$\begingroup$

as weights?

What's the advantage or disadvantage?

$\endgroup$
  • $\begingroup$ Weight inversely proportional to variance minimizes the variance of the resulting estimate. $\endgroup$ – Glen_b Mar 12 '17 at 6:11
0
$\begingroup$

Under assumptions of independent, normally distributed residuals, weighting at 1/variance corresponds to a maximum likelihood estimator for the model, so it's easier to justify from theory (subject to the usual quibbles about MLE).

$\endgroup$

Your Answer

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

Not the answer you're looking for? Browse other questions tagged or ask your own question.