Raw Message

Message-ID: <20180324142625.63291098@ECM-DTC-716.uniwa.uwa.edu.au>
Date: 2018-03-24T06:26:25Z
From: Berwin A Turlach
Subject: Question about gls
In-Reply-To: <CA+XOvOTtW8L_yVpteRdsVg0iPS1v902jwYAf1TiR95H4Pt=azA@mail.gmail.com>

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