jackhlj 发表于 2015-5-19 16:08
麻烦楼主抽空帮忙看看,谢谢
输出文件结果:
Output from the program FRONTIER (Version 4.1c)
instruction file = EG1-ins.txt
data file = wy-dta.txt
Error Components Frontier (see B&C 1992)
The model is a production function
The dependent variable is not logged
the ols estimates are :
coefficient standard-error t-ratio
beta 0 0.14203342E+01 0.41643792E+01 0.34106746E+00
beta 1 0.10365761E+01 0.22976308E+00 0.45114999E+01
beta 2 -0.10690554E-01 0.27271486E-01 -0.39200483E+00
sigma-squared 0.98850658E+02
log likelihood function = -0.10989190E+03
the estimates after the grid search were :
beta 0 0.13088385E+02
beta 1 0.10365761E+01
beta 2 -0.10690554E-01
sigma-squared 0.22510899E+03
gamma 0.95000000E+00
mu 0.00000000E+00
eta 0.00000000E+00
iteration = 0 func evals = 20 llf = -0.90209958E+02
0.13088385E+02 0.10365761E+01-0.10690554E-01 0.22510899E+03 0.95000000E+00
0.00000000E+00 0.00000000E+00
gradient step
iteration = 5 func evals = 54 llf = -0.85339346E+02
0.13356170E+02 0.10129936E+01 0.22246698E-01 0.22511867E+03 0.98409327E+00
0.92196049E-01 0.25852251E-01
iteration = 10 func evals = 147 llf = -0.82901729E+02
0.21790533E+02 0.59239401E+00 0.33249589E-01 0.22525827E+03 0.98692210E+00
0.30494366E+01 0.28275438E-01
iteration = 15 func evals = 261 llf = -0.81527212E+02
0.23672592E+02 0.43118528E+00 0.64879377E-01 0.22601884E+03 0.98829961E+00
0.17036200E+02 0.12777589E-02
iteration = 20 func evals = 360 llf = -0.81523438E+02
0.23356010E+02 0.44508389E+00 0.64580516E-01 0.22425055E+03 0.98826658E+00
0.16980381E+02 0.68442566E-03
iteration = 25 func evals = 490 llf = -0.81484303E+02
0.21973457E+02 0.52160816E+00 0.58346609E-01 0.18496634E+03 0.98563356E+00
0.17117989E+02 0.30896874E-02
iteration = 30 func evals = 616 llf = -0.81445940E+02
0.21909288E+02 0.51548845E+00 0.57409832E-01 0.15740758E+03 0.98290798E+00
0.17506551E+02 0.54624234E-02
iteration = 35 func evals = 756 llf = -0.81423480E+02
0.23037498E+02 0.44420318E+00 0.63379874E-01 0.16373624E+03 0.98454603E+00
0.17731204E+02 0.26839692E-03
iteration = 40 func evals = 889 llf = -0.81337160E+02
0.23993414E+02 0.44117826E+00 0.60329102E-01 0.14622072E+03 0.98176678E+00
0.18597461E+02 0.53140242E-02
iteration = 45 func evals = 1026 llf = -0.81317825E+02
0.23568575E+02 0.46945999E+00 0.60266901E-01 0.14045209E+03 0.98090037E+00
0.19248504E+02 0.49554788E-02
search failed. fn val indep of search direction
iteration = 47 func evals = 1046 llf = -0.81317825E+02
0.23568332E+02 0.46950702E+00 0.60261795E-01 0.14051919E+03 0.98090599E+00
0.19245205E+02 0.49592835E-02
the final mle estimates are :
coefficient standard-error t-ratio
beta 0 0.23568332E+02 0.40854653E+01 0.57688244E+01
beta 1 0.46950702E+00 0.22511755E+00 0.20856083E+01
beta 2 0.60261795E-01 0.31900594E-01 0.18890493E+01
sigma-squared 0.14051919E+03 0.88795105E+02 0.15825106E+01
gamma 0.98090599E+00 0.13849789E-01 0.70824616E+02
mu 0.19245205E+02 0.62146733E+01 0.30967365E+01
eta 0.49592835E-02 0.30760535E-01 0.16122228E+00
log likelihood function = -0.81317825E+02
LR test of the one-sided error = 0.57148152E+02
with number of restrictions = 3
[note that this statistic has a mixed chi-square distribution]
number of iterations = 47
(maximum number of iterations set at : 100)
number of cross-sections = 10
number of time periods = 3
total number of observations = 30
thus there are: 0 obsns not in the panel
covariance matrix :
0.16691026E+02 -0.65851734E+00 0.23925816E-01 0.34135882E+02 0.55510067E-02
0.83039270E+01 -0.95703625E-04
-0.65851734E+00 0.50677910E-01 -0.47429581E-02 -0.57687400E+01 -0.11006204E-02
-0.30279590E+00 0.30309530E-02
0.23925816E-01 -0.47429581E-02 0.10176479E-02 0.73445554E+00 0.15523539E-03
0.78706208E-01 -0.81372948E-03
0.34135882E+02 -0.57687400E+01 0.73445554E+00 0.78845706E+04 0.11041159E+01
-0.13174489E+03 -0.68084629E+00
0.55510067E-02 -0.11006204E-02 0.15523539E-03 0.11041159E+01 0.19181665E-03
-0.15666062E-01 -0.13779924E-03
0.83039270E+01 -0.30279590E+00 0.78706208E-01 -0.13174489E+03 -0.15666062E-01
0.38622164E+02 -0.66869110E-01
-0.95703625E-04 0.30309530E-02 -0.81372948E-03 -0.68084629E+00 -0.13779924E-03
-0.66869110E-01 0.94621053E-03
technical efficiency estimates :
efficiency estimates for year 1 :
firm eff.-est.
1 0.36839846E+00
2 0.32888659E+00
3 0.94797341E+00
4 0.33243946E+00
5 0.44502373E+00
6 0.50962225E+00
7 0.68885235E+00
8 0.91991779E+00
9 0.74365124E-01
10 0.13863328E+00
mean eff. in year 1 = 0.47541124E+00
efficiency estimates for year 2 :
firm eff.-est.
1 0.37152299E+00
2 0.33220659E+00
3 0.94823079E+00
4 0.33574189E+00
5 0.44776920E+00
6 0.51204815E+00
7 0.69039160E+00
8 0.92031396E+00
9 0.78944246E-01
10 0.14289447E+00
mean eff. in year 2 = 0.47800639E+00
efficiency estimates for year 3 :
firm eff.-est.
1 0.37463207E+00
2 0.33551017E+00
3 0.94848689E+00
4 0.33902797E+00
5 0.45050109E+00
6 0.51446205E+00
7 0.69192323E+00
8 0.92070816E+00
9 0.83500715E-01
10 0.14713458E+00
mean eff. in year 3 = 0.48058869E+00
summary of panel of observations:
(1 = observed, 0 = not observed)
t: 1 2 3
n
1 1 1 1 3
2 1 1 1 3
3 1 1 1 3
4 1 1 1 3
5 1 1 1 3
6 1 1 1 3
7 1 1 1 3
8 1 1 1 3
9 1 1 1 3
10 1 1 1 3
10 10 10 30
这是我给你运行出来的结果,你的问题在于区域只有十个,分时期时你没改变区域编号,直接编到30号,另外你的引导文件也有问题,我给你贴上去。
数据文件:
1 1 10.40 5.40 65.10
2 1 11.70 17.10 160.40
3 1 31.60 17.10 53.20
4 1 14.60 26.40 231.80
5 1 18.30 30.80 114.50
6 1 17.60 20.70 80.70
7 1 30.80 20.90 232.80
8 1 41.40 33.90 161.10
9 1 2.30 12.00 78.60
10 1 3.10 5.40 13.30
1 2 11.00 6.60 75.10
2 2 14.60 16.20 177.30
3 2 34.20 19.00 50.70
4 2 18.20 26.60 263.10
5 2 21.60 32.00 140.80
6 2 20.60 22.80 98.90
7 2 35.30 24.20 267.70
8 2 46.50 35.30 168.40
9 2 2.60 12.90 101.60
10 2 3.90 5.40 16.10
1 3 13.10 5.80 88.70
2 3 16.10 17.70 218.40
3 3 36.90 22.70 52.50
4 3 20.60 33.00 317.10
5 3 24.00 36.50 171.00
6 3 22.50 19.30 124.50
7 3 39.60 25.80 312.60
8 3 51.90 35.90 193.10
9 3 3.20 10.80 126.40
10 3 4.50 4.50 21.50
引导文件:
1 1=ERROR COMPONENTS MODEL, 2=TE EFFECTS MODEL
wy-dta.txt DATA FILE NAME
wy-out.txt OUTPUT FILE NAME
1 1=PRODUCTION FUNCTION, 2=COST FUNCTION
n LOGGED DEPENDENT VARIABLE (Y/N)
10 NUMBER OF CROSS-SECTIONS
3 NUMBER OF TIME PERIODS
30 NUMBER OF OBSERVATIONS IN TOTAL
2 NUMBER OF REGRESSOR VARIABLES (Xs)
y MU (Y/N) [OR DELTA0 (Y/N) IF USING TE EFFECTS MODEL]
y ETA (Y/N) [OR NUMBER OF TE EFFECTS REGRESSORS (Zs)]
n STARTING VALUES (Y/N)
IF YES THEN BETA0
BETA1 TO
BETAK
SIGMA SQUARED
GAMMA
MU [OR DELTA0
ETA DELTA1 TO
DELTAP]
NOTE: IF YOU ARE SUPPLYING STARTING VALUES
AND YOU HAVE RESTRICTED MU [OR DELTA0] TO BE
ZERO THEN YOU SHOULD NOT SUPPLY A STARTING
VALUE FOR THIS PARAMETER.