Thanks for the reply, but I looked in each element of names(lme_res2) but did
not find the value I was looking for. Is there a way to output the object
contents to a text file? I can then lift the value with text manipulation.
Obviously, other solutions are welcome, too.
Shin
----- Original Message -----
From: Moshe Olshansky <m_olshansky at yahoo.com>
Date: Monday, November 26, 2007 10:50 pm
Subject: Re: [R] lme object manipulation
To: SHIN LIN <slina at jhmi.edu>
> There may be a better way of doing this, but one
> possibility is to look at names(lme_res2) and then if
> something whose name is x may be of interest to you,
> look at lme_res2$x.
>
> --- SHIN LIN <slina at jhmi.edu> wrote:
>
> > Hello:
> >
> > I have an lme object, say lme_res2, which was
> > generated using the varIdent. I'm trying to extract
> > the double 1.532940 from the object, but I can't
> > find it by attributes(lme_res2) or
> > attributes(summary(lme_res2)). How can I pull it
> > out (so that I can save it to another variable)?
> > Thanks.
> >
> > Shin
> >
> > Linear mixed-effects model fit by REML
> > Data: dat
> > Log-restricted-likelihood: 22.3213
> > Fixed: express1 ~ race * thaps + race * tuni
> > (Intercept) race thaps tuni
> > race:thaps race:tuni
> > 7.00034057 -0.02675202 -0.05687972 0.01265845
> > -0.03859781 -0.11248439
> > Random effects:
> > Formula: ~1 | fam
> > (Intercept)
> > StdDev: 0.1131353
> > Formula: ~1 | twin %in% fam
> > (Intercept)
> > StdDev: 0.1740635
> > Formula: ~1 | thaw %in% twin %in% fam
> > (Intercept) Residual
> > StdDev: 0.1177579 0.1182267
> > Variance function:
> > Structure: Different standard deviations per
> > stratum
> > Formula: ~1 | race
> > Parameter estimates:
> > 0 1
> > 1.000000 1.532940
> > Number of Observations: 218
> > Number of Groups:
> > fam twin %in% fam thaw
> > %in% twin %in% fam
> > 33 66
> > 102
> >
> > ______________________________________________
> > R-help at r-project.org mailing list
> >
> > PLEASE do read the posting guide
> >
> > and provide commented, minimal, self-contained,
> > reproducible code.
> >