Stata The Stata listserver
[Date Prev][Date Next][Thread Prev][Thread Next][Date index][Thread index]

st: Gllamm question


From   "Chih-Mao Hsieh" <hsieh@wustl.edu>
To   <statalist@hsphsun2.harvard.edu>
Subject   st: Gllamm question
Date   Sat, 1 Nov 2003 12:18:51 -0600

Hello all,

I'm running gllamm and I'm trying to figure out what/how to change
w.r.t. quadrature points (log file attached).  Is the option simply
"nip()" and if so, how do I know the potential candidates for what goes
inside the ()?

Thanks in advance,
Chihmao.


-----------------------------------------------------
Chihmao Hsieh
John M. Olin School of Business
Washington University
Box 1133, One Brookings Drive
St. Louis, MO 63130
Email:  hsieh@olin.wustl.edu
http://students.olin.wustl.edu/~hsieh

------------------------------------------------------
       log:  m:\gllammrun.txt
  log type:  text
 opened on:  31 Oct 2003, 21:52:43

. use c:\tempcmh\finalfileforregressionanalysis ;

. keep citing nclass v6 v9 v12 v15 v18 v21 v24 v27 ;

. ren v6 pattern1 ;

. ren v9 pattern2 ;

. ren v12 pattern3 ;

. ren v15 pattern4 ;

. ren v18 pattern5 ;

. ren v21 pattern6 ;

. ren v24 pattern7 ;

. ren v27 pattern8 ;

. keep if nclass==73 ;
(1750479 observations deleted)

. reshape long pattern, i(citing) j(type) ;
(note: j = 1 2 3 4 5 6 7 8)
pattern4:  361 values would be changed; not changed
pattern5:  4611 values would be changed; not changed
pattern6:  4611 values would be changed; not changed
pattern7:  7427 values would be changed; not changed
pattern8:  9490 values would be changed; not changed

Data                               wide   ->   long
------------------------------------------------------------------------
-----
Number of obs.                    24682   ->  197456
Number of variables                  10   ->       4
j variable (8 values)                     ->   type
xij variables:
         pattern1 pattern2 ... pattern8   ->   pattern
------------------------------------------------------------------------
-----

. tab type, gen(t) ;

       type |      Freq.     Percent        Cum.
------------+-----------------------------------
          1 |     24,682       12.50       12.50
          2 |     24,682       12.50       25.00
          3 |     24,682       12.50       37.50
          4 |     24,682       12.50       50.00
          5 |     24,682       12.50       62.50
          6 |     24,682       12.50       75.00
          7 |     24,682       12.50       87.50
          8 |     24,682       12.50      100.00
------------+-----------------------------------
      Total |    197,456      100.00

. eq load: t1-t8 ;

. gllamm pattern t1-t8, i(citing) eqs(load) nocons f(poisson) adapt
trace ;

General model information
------------------------------------------------------------------------
------

dependent variable:         pattern
family:                     poiss
link:                       log
equation for fixed effects  t1 t2 t3 t4 t5 t6 t7 t8

Random effects information for 2 level model
------------------------------------------------------------------------
------



***level 2 (citing) equation(s):
   (1 random effect(s))


   lambdas for random effect 1
   cit1_1l: t2 t3 t4 t5 t6 t7 t8
   standard deviation for random effect 1
   cit1_1 : t1

number of level 1 units = 197456
number of level 2 units = 24682

Initial values for fixed effects

note: you are responsible for interpretation of non-count dep. variable

Iteration 0:   log likelihood =  -21109368
Iteration 1:   log likelihood =  -12875305
Iteration 2:   log likelihood =  -12848410
Iteration 3:   log likelihood =  -12848293
Iteration 4:   log likelihood =  -12848293

Poisson regression                                Number of obs   =
197456
                                                  Wald chi2(8)    =
7.860e+07
Log likelihood =  -12848293                       Prob > chi2     =
0.0000

------------------------------------------------------------------------
------
     pattern |      Coef.   Std. Err.      z    P>|z|     [95% Conf.
Interval]
-------------+----------------------------------------------------------
------
          t1 |  -2.061391   .0178417  -115.54   0.000     -2.09636
-2.026422
          t2 |  -1.183642   .0115036  -102.89   0.000    -1.206189
-1.161095
          t3 |  -2.022445   .0174976  -115.58   0.000     -2.05674
-1.98815
          t4 |  -.1825347   .0069734   -26.18   0.000    -.1962024
-.168867
          t5 |   .4198422   .0051599    81.37   0.000      .409729
.4299554
          t6 |   .4198422   .0051599    81.37   0.000      .409729
.4299554
          t7 |   1.837066   .0025404   723.15   0.000     1.832087
1.842045
          t8 |   4.885975   .0005531  8833.19   0.000     4.884891
4.88706
------------------------------------------------------------------------
------
------------------------------------------------------------------------
------


start running on 31 Oct 2003 at 21:58:42

Running adaptive quadrature
------------------------------------------------------------------------
------
Iteration 0 of adaptive quadrature:
Initial parameters:

      pattern:   pattern:   pattern:   pattern:   pattern:   pattern:
pattern:   pattern:   cit1_1l:
           t1         t2         t3         t4         t5         t6
t7         t8         t2
y1  -2.061391  -1.183642  -2.022445  -.1825347   .4198422   .4198422
1.837066   4.885975          1

      cit1_1l:   cit1_1l:   cit1_1l:   cit1_1l:   cit1_1l:   cit1_1l:
cit1_1:
           t3         t4         t5         t6         t7         t8
t1
y1          1          1          1          1          1          1
.5

Updated log likelihood:
-5.146e+06 -5.232e+06 -5.210e+06 -5.188e+06 -5.165e+06 -5.137e+06
-5.115e+06 -5.108e+06 -5.105e+06 -5.102e+06 -5.099e+06 -5.097e+06
-5.095e+06 -5.094e+06 -5.092e+06 -5.090e+06 -5.089e+06 -5.088e+06
-5.086e+06 -5.085e+06 -5.084e+06 -5.083e+06 -5.082e+06 -5.080e+06
-5.079e+06 -5.078e+06 -5.077e+06 -5.076e+06 -5.075e+06 -5.074e+06
-5.072e+06 -5.071e+06 -5.070e+06 -5.069e+06 -5.068e+06 -5.067e+06
-5.066e+06 -5.065e+06 -5.064e+06 -5.063e+06 -5.062e+06 -5.061e+06
-5.060e+06 -5.059e+06 -5.058e+06 -5.057e+06 -5.056e+06 -5.055e+06
-5.054e+06 -5.053e+06 -5.052e+06 -5.051e+06 -5.050e+06 -5.049e+06
-5.048e+06 -5.047e+06 -5.046e+06 -5.045e+06 -5.044e+06 -5.044e+06
-5.043e+06 -5.042e+06 -5.041e+06 -5.040e+06 -5.039e+06 -5.038e+06
-5.037e+06 -5.036e+06 -5.035e+06 -5.035e+06 -5.034e+06 -5.033e+06
-5.032e+06 -5.031e+06 -5.030e+06 -5.029e+06 -5.028e+06 -5.027e+06
-5.027e+06 -5.026e+06 -5.025e+06 -5.024e+06 -5.023e+06 -5.022e+06
-5.021e+06 -5.020e+06 -5.020e+06 -5.019e+06 -5.018e+06 -5.017e+06
-5.016e+06 -5.015e+06 -5.014e+06 -5.013e+06 -5.013e+06 -5.012e+06
-5.011e+06 -5.010e+06 -5.009e+06 -5.008e+06 -5.008e+06 -5.007e+06
-5.006e+06 -5.005e+06 -5.004e+06 -5.003e+06 -5.003e+06 -5.002e+06
-5.001e+06 -5.000e+06 -4.999e+06 -4.998e+06 -4.998e+06 -4.997e+06
-4.996e+06 -4.995e+06 -4.994e+06 -4.994e+06 -4.993e+06 Convergence not
achieved: try with more quadratu
> re points
finish running on 31 Oct 2003 at 22:30:06



*
*   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/



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