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: [email protected]
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/
