st: post estimation in gllamm again

[Michele Guerin <mguerin@ovc.uoguelph.ca>]

Thanks Caleb, I should have included more information.  Perhaps it is a
df problem.  The random effects in my full model are also very small. 
Here is my code and output:


NULL MODEL:

gllamm status, fam(bin) link(logit) i(e_lufsitecode e_farm) adapt
 
number of level 1 units = 792
number of level 2 units = 83
number of level 3 units = 33
 
Condition Number = 1.2407365
 
gllamm model
 
log likelihood = -446.21472
 
------------------------------------------------------------------------------
      status |      Coef.   Std. Err.      z    P>|z|     [95% Conf.
Interval]
-------------+----------------------------------------------------------------
       _cons |  -1.240137   .1827224    -6.79   0.000    -1.598266  
-.8820074
------------------------------------------------------------------------------
 
Variances and covariances of random effects
-----------------------------------------------------------------------------
 
***level 2 (e_lufsitecode)
 
    var(1): 1.431e-15 (1.463e-08)
 
***level 3 (e_farm)
 
    var(1): .60448629 (.26422999)
-----------------------------------------------------------------------------

. estat ic

------------------------------------------------------------------------------
       Model |    Obs    ll(null)   ll(model)     df         
AIC         BIC
-------------+----------------------------------------------------------------
           . |     33           .   -446.2147      3     898.4294    
902.919
------------------------------------------------------------------------------
 

FULL MODEL:

xi: gllamm status wNT wOT i.bis1B sas1 scumdd403cen10 scumdd403censq10
snumt483bin ageP size_flock, fam(bin) link(logit) i(e_lufsitecode
e_farm) adapt
i.bis1B           _Ibis1B_0-2         (naturally coded; _Ibis1B_0
omitted)
 
number of level 1 units = 694
number of level 2 units = 80
number of level 3 units = 31
 
Condition Number = 149248.47
 
gllamm model
 
log likelihood = -329.04972
 
------------------------------------------------------------------------------
      status |      Coef.   Std. Err.      z    P>|z|     [95% Conf.
Interval]
-------------+----------------------------------------------------------------
         wNT |  -.3849298   .2427759    -1.59   0.113    -.8607617   
.0909022
         wOT |  -1.101316   .4090949    -2.69   0.007    -1.903128  
-.2995053
   _Ibis1B_2 |   .6672017   .2577098     2.59   0.010     .1620998   
1.172304
        sas1 |   1.440306   .2811731     5.12   0.000     .8892169   
1.991395
scumdd40~n10 |   .0828534   .0351455     2.36   0.018     .0139694   
.1517374
scumdd40~q10 |   .0016047   .0004363     3.68   0.000     .0007496   
.0024599
 snumt483bin |  -1.153253   .3989543    -2.89   0.004    -1.935189  
-.3713174
        ageP |   .1714664   .0407183     4.21   0.000       .09166   
.2512727
  size_flock |   .0000608   .0000242     2.52   0.012     .0000134   
.0001081
       _cons |  -9.315843   1.543667    -6.03   0.000    -12.34137  
-6.290311
------------------------------------------------------------------------------
 
Variances and covariances of random effects
-----------------------------------------------------------------------------
 
***level 2 (e_lufsitecode)
 
    var(1): 1.865e-17 (1.452e-09)
 
***level 3 (e_farm)
 
    var(1): 3.769e-15 (1.984e-08)
-----------------------------------------------------------------------------

. est store N

. estat ic

------------------------------------------------------------------------------
       Model |    Obs    ll(null)   ll(model)     df         
AIC         BIC
-------------+----------------------------------------------------------------
           N |     31           .   -329.0497     12     682.0994   
699.3073
------------------------------------------------------------------------------

Michele
*
*   For searches and help try:
*   http://www.stata.com/support/faqs/res/findit.html
*   http://www.stata.com/support/statalist/faq
*   http://www.ats.ucla.edu/stat/stata/