[R-sig-ME] How to find ACF, PACF, Sample Variance-Covariance Matrix of Random-Effects?
Ben Bolker
bbo|ker @end|ng |rom gm@||@com
Tue Oct 11 02:27:56 CEST 2022
Can you give an example please?
The covariance matrix is typically (for nlme and other packages that
follow the same conventions) extracted with VarCorr(). As described in
my other e-mail to the list, this is not *quite* the same as the full
covariance matrix of the random-effects vector, but rather the
covariance matrix for a single grouping variable/block. VarCorr()
returns a list. So e.g. if
library(lme4)
fm1 <- lmer(Reaction ~ Days + (Days|Subject), sleepstudy)
VarCorr(fm1)$Subject gives a 2x2 covariance matrix
cov2cor() will convert this to a correlation matrix.
Or, for lmer fits at least, the correlation is already present as an
attribute:
attr(VarCorr(fm1)$Subject, "correlation")
On 2022-10-07 9:43 a.m., Sun, John wrote:
> Dear All,
>
> I am writing to ask how to compute the (partial) autocorrelation of random-effects versus lag, and sample-variance covariance matrix of the random-effects given some data without specification of some model or structure on the random-effects covariance matrix.
>
> Best regards,
> John
>
> _______________________________________________
> R-sig-mixed-models using r-project.org mailing list
> https://stat.ethz.ch/mailman/listinfo/r-sig-mixed-models
--
Dr. Benjamin Bolker
Professor, Mathematics & Statistics and Biology, McMaster University
Director, School of Computational Science and Engineering
(Acting) Graduate chair, Mathematics & Statistics
> E-mail is sent at my convenience; I don't expect replies outside of
working hours.
More information about the R-sig-mixed-models
mailing list