[Limdep Nlogit List] Round 2: Fixed effects model: Error 682
William Greene
wgreene at stern.nyu.edu
Wed Jan 23 05:27:56 EST 2008
Dear Pavlos. This looks to me like a malfunction of the program. I will
have to do some work to find out. I will let you know.
Regards,
Bill Greene
----- Original Message -----
From: "Pavlos C. Symeou" <p.symeou at jbs.cam.ac.uk>
To: limdep at limdep.itls.usyd.edu.au
Sent: Monday, January 14, 2008 1:28:10 PM (GMT-0500) America/New_York
Subject: [Limdep Nlogit List] Round 2: Fixed effects model: Error 682
Dear Profs Hofler and Greene,
thank you for your feedback. I have followed your instructions but I am
still encountering the same problem. "Error 682: Did not find matching
previous nonpanel FRONTIER model."
Let me provide you some more information about my commands and
respective output:
Starting with the following commands to estimate a True FEM
Frontier ; Lhs = ly1; Rhs = one, xs $
Frontier ; Lhs = ly1; Rhs = one, xs ; Model=T
; FEM ; Pds = ni $
Limdep generates the following output:
Nonlinear Estimation of Model Parameters
Method=Newton; Maximum iterations=100
Convergence criteria: max|dB| .1000D-08, dF/F= .1000D-08, g<H>g=
.1000D-05
Normal exit from iterations. Exit status=0.
+---------------------------------------------+
| FIXED EFFECTS Frontr Model |
| Maximum Likelihood Estimates |
| Model estimated: Jan 14, 2008 at 05:52:17PM.|
| Dependent variable LY1 |
| Weighting variable None |
| Number of observations 442 |
| Iterations completed 8 |
| Log likelihood function -468.8782 |
| Number of parameters 75 |
| Info. Criterion: AIC = 2.46099 |
| Finite Sample: AIC = 2.53146 |
| Info. Criterion: BIC = 3.15521 |
| Info. Criterion:HQIC = 2.73481 |
| Unbalanced panel has 58 individuals. |
| Bypassed 0 groups with inestimable a(i). |
| Half normal stochastic frontier |
| Sigma( u) (1 sided) = 2.05761 |
| Sigma( v) (symmetric)= .14032 |
+---------------------------------------------+
+--------+--------------+----------------+--------+--------+----------+
|Variable| Coefficient | Standard Error |b/St.Er.|P[|Z|>z]| Mean of X|
+--------+--------------+----------------+--------+--------+----------+
---------+Production / Cost parameters
LX1 | 3.44276913 .02884619 119.349 .0000 17.2717635
LX1LX2 | -.07250239 .00427867 -16.945 .0000 133.366836
LX1LY2S | .13220888 .00284707 46.437 .0000 -26.2072718
LX1LY3S | -.07142750 .00341732 -20.902 .0000 -42.5942353
LX1SQ | -.07445498 .00120234 -61.925 .0000 301.198362
LX2 | 1.17689041 .05923370 19.869 .0000 7.66565044
LX2LY2S | -.05305911 .00383705 -13.828 .0000 -12.0554763
LX2LY3S | .01823538 .00593461 3.073 .0021 -19.2147980
LX2SQ | -.00707330 .00662746 -1.067 .2858 59.9286892
LY2S | -1.94640051 .04119240 -47.251 .0000 -1.54591389
LY2SLY3S| .09366029 .00502577 18.636 .0000 6.18072655
LY2SSQ | -.07503178 .00198996 -37.705 .0000 5.52582271
LY3S | 1.39172602 .04834666 28.786 .0000 -2.48318078
LY3SSQ | -.01018299 .00450384 -2.261 .0238 9.15073948
T | -.07826962 .00439862 -17.794 .0000 8.56787330
---------+Variance parameter for v +/- u
Sigma | 2.06238690 .01144773 180.157 .0000
---------+Asymmetry parameter, lambda
Lambda | 14.6634877 1.21611828 12.058 .0000
I then try to estimate the Normal-Truncated Normal Model. I try
different variations. Namely, with Heterogeneity appearing in the
production function and therefore only the first FRONTIER includes ";
Model = T" command. Or with heterogeneity to the mean of the truncated
distribution and therefore both FRONTIER lines contain the "; Model = T"
command as below.
Frontier ; Lhs = ly1; Rhs = one, xs; Rh2 = one, zs; Model=T $
Frontier ; Lhs = ly1; Rhs = one, xs; Rh2 = one, zs; Model=T
; FEM ; Pds = ni $
Nevertheless, in either case Limdep gives me the "Error 682: Did not
find matching previous nonpanel FRONTIER model." Below I give you the
output when the first FRONTIER line of the above command is estimated.
The second FRONTIER command creates "Error 682". As you can see in the
first line, estimation stops when the Maximum iterations are reached. I
changed my model so that it would exit with normal status but I still
got Error 682.
Maximum iterations reached. Exit iterations with status=1.
+---------------------------------------------+
| Limited Dependent Variable Model - FRONTIER |
| Maximum Likelihood Estimates |
| Model estimated: Jan 14, 2008 at 06:05:56PM.|
| Dependent variable LY1 |
| Weighting variable None |
| Number of observations 442 |
| Iterations completed 101 |
| Log likelihood function -505.5538 |
| Number of parameters 24 |
| Info. Criterion: AIC = 2.39617 |
| Finite Sample: AIC = 2.40268 |
| Info. Criterion: BIC = 2.61832 |
| Info. Criterion:HQIC = 2.48379 |
| Variances: Sigma-squared(v)= .00003 |
| Sigma-squared(u)= 2.91238 |
| Sigma(v) = .00589 |
| Sigma(u) = 1.70657 |
| Sigma = Sqr[(s^2(u)+s^2(v)]= 1.70658 |
| Stochastic Production Frontier, e=v-u. |
+---------------------------------------------+
+--------+--------------+----------------+--------+--------+----------+
|Variable| Coefficient | Standard Error |b/St.Er.|P[|Z|>z]| Mean of X|
+--------+--------------+----------------+--------+--------+----------+
---------+Primary Index Equation for Model
Constant| -24.3920960 3.54554615 -6.880 .0000
LX1 | 2.71002995 .50431701 5.374 .0000 17.2717635
LX1LX2 | .06985315 .03289810 2.123 .0337 133.366836
LX1LY2S | .15066697 .02886164 5.220 .0000 -26.2072718
LX1LY3S | -.08558477 .02372664 -3.607 .0003 -42.5942353
LX1SQ | -.08695433 .01452261 -5.988 .0000 301.198362
LX2 | 1.04718944 .35411933 2.957 .0031 7.66565044
LX2LY2S | -.11665260 .04469894 -2.610 .0091 -12.0554763
LX2LY3S | .08466289 .05344673 1.584 .1132 -19.2147980
LX2SQ | -.15922215 .04018984 -3.962 .0001 59.9286892
LY2S | -1.66157925 .31610784 -5.256 .0000 -1.54591389
LY2SLY3S| .11670543 .03776725 3.090 .0020 6.18072655
LY2SSQ | -.06411783 .01844242 -3.477 .0005 5.52582271
LY3S | .88360417 .33840414 2.611 .0090 -2.48318078
LY3SSQ | -.04056935 .01874473 -2.164 .0304 9.15073948
T | -.01701402 .02751700 -.618 .5364 8.56787330
---------+Offset [mean=mu(i)] parameters in one sided error
Constant| 4.56462131 28.0437363 .163 .8707 1.00000000
LIBERA | .25402691 1.57188632 .162 .8716 .56108597
NRA | .81004648 4.99042802 .162 .8711 .41402715
OPENC | -.03383148 .20779730 -.163 .8707 113.253949
POLCONII| -6.25797571 38.4764904 -.163 .8708 .28912146
TRINITY | .38287561 .44558410 .859 .3902 .15837104
---------+Variance parameters for compound error
Lambda | 289.839164 1772.43389 .164 .8701
Sigma | 1.70657944 .18946840 9.007 .0000
_______________________________________________
Limdep site list
Limdep at limdep.itls.usyd.edu.au
http://limdep.itls.usyd.edu.au
--
Professor William Greene
Department of Economics
Stern School of Business
New York University
44 West 4th St., Rm. 7-78
New York, NY 10012
http://www.stern.nyu.edu/~wgreene
212.998.0876
More information about the Limdep
mailing list