Statalist


[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

st: issue using starting values in gllamm


From   "Fiona Gibson" <gibsof01@student.uwa.edu.au>
To   <statalist@hsphsun2.harvard.edu>
Subject   st: issue using starting values in gllamm
Date   Tue, 21 Apr 2009 13:04:57 +0800

Dear Stata list members

I am estimating a 2 level mixed response model using panel data. The
response is dichotomous (0,1) and a function of observed dummy variables,
with direct and indirect effects, and ordered categorical latent variables
(Chapter 6 GLLAMM Manual).  Using ?gllamm- (in Stata 10.1) I have run the
following model.  

eq id1: op1 op2 op3 yvar
eq id2: op4 op5 op6 op7 yvar
eq id3: op8 op9 op10 op11 yvar
eq f1: male ter nochild
eq f2: male ter nochild
eq f3: male ter nochild

gllamm resp  yvar  bid1 ter2 male2 nochild2 op2 op3 op5 op6 op7 op9 op10
op11 , nip(8) nrf(3)  i(id) eqs(id1 id2 id3) geqs(f1 f2 f3) link(ologit
ologit ologit logit) family(binomial binomial binomial binomial) lv(var)
fv(var) from(ex2) trace adapt

Equations id1-id3 are the measurement models, f1-f3 the ?exposure? models.

The model estimated and appeared to converge (results reported below).  To
check the robustness of parameters at different ?nip()-, I retrieved the
parameter matrix (ex2) to use as starting values. However, after using
?from(ex2)- to input the starting values, at iteration 0 it reports:

Updated log likelihood:
         . Log-likelihood cannot be computed
finish running on 21 Apr 2009 at 09:42:00

This occurs even if -nip(8)- is used. Any idea why that may be so?

Your help is appreciated.
  
Regards
Fiona Gibson
PhD Candidate
School of Agricultural and Resource Economics
University of Western Australia
Email: gibsof01@student.uwa.edu.au


Model results
number of level 1 units = 5688
number of level 2 units = 474
 
Condition Number = 1334.7088
 
gllamm model
 
log likelihood = -5522.7565
 
----------------------------------------------------------------------------
--
        resp |      Coef.   Std. Err.      z    P>|z|     [95% Conf.
Interval]
-------------+--------------------------------------------------------------
--
resp         |
        yvar |  -.9929185   .2627392    -3.78   0.000    -1.507878  
-.4779591
        bid1 |  -.0087789   .0013141    -6.68   0.000    -.0113545  
-.0062032
        ter2 |   .5641199   .2353094     2.40   0.017      .102922   
1.025318
       male2 |  -.3873411   .2378501    -1.63   0.103    -.8535186   
.0788365
    nochild2 |   .4173029   .2298836     1.82   0.069    -.0332606   
.8678665
         op2 |    .301384   .1795849     1.68   0.093    -.0505958   
.6533639
         op3 |   .4976204   .2281599     2.18   0.029     .0504352   
.9448056
         op5 |  -.3044933   .1789783    -1.70   0.089    -.6552844   
.0462978
         op6 |  -.1403718   .1687842    -0.83   0.406    -.4711828   
.1904391
         op7 |  -.5554956   .1942089    -2.86   0.004    -.9361381  
-.1748532
         op9 |   -.410213   .1346565    -3.05   0.002    -.6741349  
-.1462911
        op10 |    2.12092   .1470227    14.43   0.000     1.832761   
2.409079
        op11 |  -.5361621   .5208446    -1.03   0.303    -1.556999   
.4846744
-------------+--------------------------------------------------------------
--
_cut11       |
       _cons |  -8.396816   .7351916   -11.42   0.000    -9.837765  
-6.955867
-------------+--------------------------------------------------------------
--
_cut12       |
       _cons |  -4.806453   .6712836    -7.16   0.000    -6.122145  
-3.490762
-------------+--------------------------------------------------------------
--
_cut13       |
       _cons |   .6496311   .6330394     1.03   0.305    -.5911034   
1.890366
-------------+--------------------------------------------------------------
--
_cut14       |
       _cons |   5.940093   .6770886     8.77   0.000     4.613024   
7.267162
-------------+--------------------------------------------------------------
--
_cut21       |
       _cons |  -6.057203    .602005   -10.06   0.000    -7.237111  
-4.877295
-------------+--------------------------------------------------------------
--
_cut22       |
       _cons |  -2.362765   .5666456    -4.17   0.000     -3.47337   
-1.25216
-------------+--------------------------------------------------------------
--
_cut23       |
       _cons |   3.128984   .5711984     5.48   0.000     2.009455   
4.248512
-------------+--------------------------------------------------------------
--
_cut24       |
       _cons |   7.178246   .5999195    11.97   0.000     6.002426   
8.354067
-------------+--------------------------------------------------------------
--
_cut31       |
       _cons |   -4.64904   .3124781   -14.88   0.000    -5.261486  
-4.036594
-------------+--------------------------------------------------------------
--
_cut32       |
       _cons |   -1.25689   .2775795    -4.53   0.000    -1.800936  
-.7128441
-------------+--------------------------------------------------------------
--
_cut33       |
       _cons |   1.441071   .2777721     5.19   0.000     .8966474   
1.985494
-------------+--------------------------------------------------------------
--
_cut34       |
       _cons |   5.005855   .3132992    15.98   0.000       4.3918    
5.61991
----------------------------------------------------------------------------
--
 
 
Variances and covariances of random effects
----------------------------------------------------------------------------
--

 
***level 2 (id)
 
    var(1): 40.248971 (4.970909)
 
    loadings for random effect 1
    op1: 1 (fixed)
    op2: 1.0893465 (.03925325)
    op3: 1.2308001 (.04589239)
    yvar: .10182486 (.03433342)
 
    cov(2,1): 27.819774 (2.9062462) cor(2,1): .77503958
 
    var(2): 32.011399 (3.5252376)
 
    loadings for random effect 2
    op4: 1 (fixed)
    op5: .90412739 (.03409143)
    op6: .98922153 (.03568108)
    op7: .81991618 (.03114715)
    yvar: .12849134 (.03833614)
 
    cov(3,1): 5.5007943 (.98454576) cor(3,1): .32871944
    cov(3,2): 4.2710618 (.83762777) cor(3,2): .28619425
 
    var(3): 6.9573726 (.84022282)
 
    loadings for random effect 3
    op8: 1 (fixed)
    op9: 1.1164339 (.06203392)
    op10: 1.0281634 (.06511839)
    op11: -.97757119 (.05905692)
    yvar: -.20807531 (.05384573)
 
 
Regressions of latent variables on covariates
----------------------------------------------------------------------------
--

 
    random effect 1 has 3 covariates:
    male: .78137295 (.63633932)
    ter: 1.218253 (.62910364)
    nochild: .21101281 (.62815516)
    random effect 2 has 3 covariates:
    male: 1.7181385 (.56427469)
    ter: 1.2200142 (.5511976)
    nochild: .44838222 (.55413639)
    random effect 3 has 3 covariates:
    male: -.22595219 (.26684788)
    ter: .48065322 (.26684636)
    nochild: .03505078 (.26306724)
----------------------------------------------------------------------------
--



*
*   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/



© Copyright 1996–2014 StataCorp LP   |   Terms of use   |   Privacy   |   Contact us   |   What's new   |   Site index