st: estimation units for gllamm random coefficients

["Bontempo, Daniel E" <deb193@ku.edu>]

Hi all -

I can see form the output below that the modeled  units of the level-1
residual is log standard deviation, and if I exponentiate it and square
it, I get the value in the output. But this does not seem to work for
the level-2 variance. Is it in a different unit?


gllamm model
 
log likelihood = -998.21604
------------------------------------------------------------------------
------
         mDx |      Coef.   Std. Err.      z    P>|z|     [95% Conf.
Interval]
-------------+----------------------------------------------------------
------
    dualCond |   7.296223   .9465117     7.71   0.000     5.441094
9.151352
      ageGrp |   2.188459   1.317523     1.66   0.097    -.3938394
4.770757
  x_cond_age |   9.415988   1.385551     6.80   0.000     6.700357
12.13162
       _cons |   9.217251   .9020915    10.22   0.000     7.449184
10.98532
------------------------------------------------------------------------
------
Variance at level 1
------------------------------------------------------------------------
------
  35.835373 (3.3563895)
 
Variances and covariances of random effects
------------------------------------------------------------------------
------
***level 2 (subjID)
 
    var(1): 14.268077 (3.7821848)
------------------------------------------------------------------------
------

. mat list e(b)

e(b)[1,6]
           mDx:        mDx:        mDx:        mDx:       lns1:
subj1:
      dualCond      ageGrp  x_cond_age       _cons       _cons
_cons
y1   7.2962233   2.1884589   9.4159876   9.2172507   1.7894677
3.7773108

. display exp( 1.7894677 )^2

35.83537


Thanks,-Daniel

*
*   For searches and help try:
*   http://www.stata.com/help.cgi?search
*   http://www.stata.com/support/statalist/faq
*   http://www.ats.ucla.edu/stat/stata/