1
$\begingroup$

I want a marginal model, ideally fit via GLS. But the normality of residuals doesn't hold. It isn't much skewed, I don't want any transformations. It's just non-normal in shape. Yet still reporting means for that case is meaningful. The GEE estimation doesn't worry about that. Would you advice me to report rather GLS with sandwich SE, or just switch to GEE with sandwich ones?

Improve this question
$\endgroup$
1
  • 1
    $\begingroup$ I don't think that the residuals need to be normally distributed to have valid inference with GLS, does it ? In particular, I don't think that the expression of the estimator's variance doesn't assume normality. $\endgroup$
    – Pohoua
    Commented Apr 14, 2022 at 0:45

0

Reset to default

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.