Re: Question on nlme
Eric Berger <[email protected]>
| Newsgroups | gmane.comp.lang.r.general |
|---|---|
| Message-ID | <CAGgJW74OEMCTchDHbHXgkpCFhVwnpz9=KLh0CPGmTreCYKXGVQ@mail.gmail.com> |
Just reading the documentation, it seems the argument
'lmeControl(opt="optim")' says to use the optim optimizer.
Then the R command ?optim describes what is done.
?optim
"General-purpose optimization based on Nelder-Mead, quasi-Newton
and conjugate-gradient algorithms. It includes an option for
box-constrained optimization and simulated annealing."
On Sat, Sep 20, 2025 at 10:55 AM sasha a <[email protected]>
wrote:
> Hello,
>
> I am using the nlme package in R:
>
> model <- lme(
> response ~ response_type + x_y + x_z - 1,
> random = ~ response_type - 1 | group,
> weights = varIdent(form = ~ 1 | response_type),
> data = data_long,
> control = lmeControl(opt = "optim")
> )
>
> Is it possible to know what estimation process this is using? For example,
> is it using EM? Or is it use Newton-Raphson with GLS/RMLE?
>
> Thanks!
>
>
> [[alternative HTML version deleted]]
>
> ______________________________________________
> [email protected] mailing list -- To UNSUBSCRIBE and more, see
> https://stat.ethz.ch/mailman/listinfo/r-help
> PLEASE do read the posting guide
> https://www.R-project.org/posting-guide.html
> and provide commented, minimal, self-contained, reproducible code.
>
[[alternative HTML version deleted]]