I'd like to use the generalized least squares (GLS) in the multivariate version. I have a response variable $\boldsymbol{Y}$ with dimension $n \times m$, where $n$ indicates the number of observations and $m$ indicates the number of components of the random vector $\boldsymbol{Y}$. Besides, I have the regressors variables with dimension $n \times p$, where $p$ indicates the number of regressors.

How is the multivariate case for the GLS? I have read https://en.wikipedia.org/wiki/Generalized_linear_model, but this version is for the univariate case, that is, $\boldsymbol{Y}$ has dimension $n \times 1$. Any reference or hints are welcome.

  • $\begingroup$ Are you looking for software implementation of theoretical discussion ob this topic? $\endgroup$ – Isabella Ghement May 1 '19 at 2:01
  • $\begingroup$ I need a book, paper or R package. $\endgroup$ – Wagner Jorge May 1 '19 at 15:47

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.