[R] standard error of variances and covariances of the random effects with LME

Douglas Bates dmbates at gmail.com
Thu Sep 29 21:11:37 CEST 2005


With lme you could but it would take a while to work it out.  There is
an approximate Hessian matrix for the parameters that determine the
variance-covariance matrix in there somewhere and if you were
sufficiently persistent you could convert that apVar component to the
scale of the variances and covariances.  I believe it is in the scale
of the logarithm of the standard deviation and Fisher's z
transformation (i.e. the hyperbolic arc tangent) of the correlation.

On 9/29/05, Doran, Harold <HDoran at air.org> wrote:
> You cannot. Also, it's not that the distribution of the random effects
> is not symmetric, but that it *may* not be symmetric, and this is an
> assumption that should be checked.
>
> -----Original Message-----
> From: r-help-bounces at stat.math.ethz.ch
> [mailto:r-help-bounces at stat.math.ethz.ch] On Behalf Of Roel de Jong
> Sent: Thursday, September 29, 2005 9:20 AM
> To: r-help at stat.math.ethz.ch
> Subject: [R] standard error of variances and covariances of the random
> effects with LME
>
> Hello,
>
> how do I obtain standard errors of variances and covariances of the
> random effects with LME comparable to those of for example MlWin? I know
> you shouldn't use them because the distribution of the estimator isn't
> symmetric blablabla, but I need a measure of the variance of those
> estimates for pooling my multiple imputation results.
>
> Regards,
>    Roel.
>
> ______________________________________________
> R-help at stat.math.ethz.ch mailing list
> https://stat.ethz.ch/mailman/listinfo/r-help
> PLEASE do read the posting guide!
> http://www.R-project.org/posting-guide.html
>
> ______________________________________________
> R-help at stat.math.ethz.ch mailing list
> https://stat.ethz.ch/mailman/listinfo/r-help
> PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
>




More information about the R-help mailing list