[R-sig-ME] Question about gls

[Berwin A Turlach]

G'day all,

On Fri, 23 Mar 2018 15:58:11 -0400
Andrzej Galecki <agalecki at umich.edu> wrote:

> One possible way ...
> 
> mSt <- fit.gls$modelStruct
> cSt <- mSt$corStruct
> coef(cSt, unconstrained = FALSE)
> 
> #       Phi
> # 0.1748786

Another possible way:


R> (tmp <- intervals(fit.gls, which="var-cov"))
Approximate 95% confidence intervals

 Correlation structure:
         lower      est.     upper
Phi 0.08692388 0.1748786 0.2601282
attr(,"label")
[1] "Correlation structure:"

 Residual standard error:
    lower      est.     upper 
0.4820339 0.5139672 0.5480159 
R> tmp$corStruct[2]
[1] 0.1748786

Cheers,

	Berwin