Open jesus-prieto opened 3 years ago
Thank you for reporting. Seems like there is something fishy here. As far as I remember, the weights
implementation is not fully-fledged yet and does not yet support weighted residuals.
Note to the dev team:
summary(lm_object)$residuals
and lm_object$residuals
differ: summary’s residuals are the weighted residuals, see ?summary.lm
vocvXX
functions need to be enabled for weights
supportweights
support (related is issue #41)
Consider an example from the help:
As far as I'm concern, coeficcients should be equal, as the estimator used to compute them is the same. Fixed effects differ too:
In addition, residuals method provides strange results.
If we fit the model by ourselves, we obtein something which seems to be ok.
But compared with the model obtain with lm, the results seem to be worse.