[ADMB Users] NLMM Model Selection

[Ben Bolker]

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

On 11-02-06 11:15 AM, dave fournier wrote:
> I'm curious to know whether other packages besides ADMB take the
> uncertainty in the
> estimation of the other parameters when calculating the estimated
> covariance matrix for the
> posterior distribution of the random effects.

 It's not explicitly stated, but it seems that lme4 does it conditional
on the estimates of the parameters.

     If grouping factor i has k levels and j random effects per level
     the ith component of the list returned by ‘ranef’ is a data frame
     with k rows and j columns.  If ‘postVar’ is ‘TRUE’ the ‘"postVar"’
     attribute is an array of dimension j by j by k.  The kth face of
     this array is a positive definite symmetric j by j matrix.  If
     there is only one grouping factor in the model the
     variance-covariance matrix for the entire random effects vector,
     conditional on the estimates of the model parameters and on the
     data will be block diagonal and this j by j matrix is the kth
     diagonal block.  With multiple grouping factors the faces of the
     ‘"postVar"’ attributes are still the diagonal blocks of this
     conditional variance-covariance matrix but the matrix itself is no
     longer block diagonal.

-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.10 (GNU/Linux)
Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org/

iEYEARECAAYFAk1O4YsACgkQc5UpGjwzenPIBQCdEYYAYecHKGB2RWzRVM/BhN99
x2UAmwas8nk3WwXcdh/OTWd+NcPuZPMK
=1BMS
-----END PGP SIGNATURE-----