发帖
雷达卡
就以下面为例吧，哪个是要估计的系数？哪个表示似然比检验时能用到的LR值？还有如果有结果显示估计的参数不明显，要把这些在模型中不明显的参数去掉后，然后再怎么继续用frontier计算修正后的结果？做模型检验时，比如零假设是伽马值为0，没有技术非效率，需要怎么检验？还望大神高手不吝赐教，拜谢拜谢！
1 1=ERROR COMPONENTS MODEL, 2=TE EFFECTS MODEL
eg1tl-d.txt DATA FILE NAME
eg1tl-o.txt OUTPUT FILE NAME
1 1=PRODUCTION FUNCTION, 2=COST FUNCTION
y LOGGED DEPENDENT VARIABLE (Y/N)
60 NUMBER OF CROSS-SECTIONS
1 NUMBER OF TIME PERIODS
60 NUMBER OF OBSERVATIONS IN TOTAL
5 NUMBER OF REGRESSOR VARIABLES (Xs)
n MU (Y/N) [OR DELTA0 (Y/N) IF USING TE EFFECTS MODEL]
n 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.
下面是结果文件：
utput from the program FRONTIER (Version 4.1c)
instruction file = eg1tl-i.txt
data file = eg1tl-d.txt
Error Components Frontier (see B&C 1992)
The model is a production function
The dependent variable is logged
the ols estimates are :
coefficient standard-error t-ratio
beta 0 0.55625525E+00 0.36338003E+00 0.15307810E+01
beta 1 0.37854397E+00 0.19384426E+00 0.19528252E+01
beta 2 0.34779921E+00 0.21214568E+00 0.16394358E+01
beta 3 -0.92959007E-01 0.45170722E-01 -0.20579482E+01
beta 4 0.30050064E-01 0.34924525E-01 0.86042870E+00
beta 5 0.84809106E-02 0.45231570E-01 0.18749981E+00
sigma-squared 0.11036714E+00
log likelihood function = -0.15857211E+02
the estimates after the grid search were :
beta 0 0.86139029E+00
beta 1 0.37854397E+00
beta 2 0.34779921E+00
beta 3 -0.92959007E-01
beta 4 0.30050064E-01
beta 5 0.84809106E-02
sigma-squared 0.19243782E+00
gamma 0.76000000E+00
mu is restricted to be zero
eta is restricted to be zero
iteration = 0 func evals = 20 llf = -0.14456850E+02
0.86139029E+00 0.37854397E+00 0.34779921E+00-0.92959007E-01 0.30050064E-01
0.84809106E-02 0.19243782E+00 0.76000000E+00
gradient step
iteration = 5 func evals = 42 llf = -0.14443212E+02
0.85314107E+00 0.37653265E+00 0.34242619E+00-0.91366437E-01 0.31419959E-01
0.88596669E-02 0.18973516E+00 0.76300678E+00
iteration = 10 func evals = 80 llf = -0.14434292E+02
0.80927823E+00 0.39187047E+00 0.36392189E+00-0.91620249E-01 0.28843663E-01
0.50747247E-02 0.18939165E+00 0.76232491E+00
iteration = 11 func evals = 85 llf = -0.14434292E+02
0.80927823E+00 0.39187047E+00 0.36392189E+00-0.91620249E-01 0.28843663E-01
0.50747247E-02 0.18939165E+00 0.76232491E+00
the final mle estimates are :
coefficient standard-error t-ratio
beta 0 0.80927823E+00 0.33438997E+00 0.24201629E+01
beta 1 0.39187047E+00 0.17841825E+00 0.21963586E+01
beta 2 0.36392189E+00 0.18391427E+00 0.19787583E+01
beta 3 -0.91620249E-01 0.41882314E-01 -0.21875642E+01
beta 4 0.28843663E-01 0.29828452E-01 0.96698491E+00
beta 5 0.50747247E-02 0.42540827E-01 0.11929069E+00
sigma-squared 0.18939165E+00 0.56787887E-01 0.33350712E+01
gamma 0.76232491E+00 0.15256058E+00 0.49968668E+01
mu is restricted to be zero
eta is restricted to be zero
log likelihood function = -0.14434292E+02
LR test of the one-sided error = 0.28458365E+01
with number of restrictions = 1
[note that this statistic has a mixed chi-square distribution]
number of iterations = 11
(maximum number of iterations set at : 100)
number of cross-sections = 60
number of time periods = 1
total number of observations = 60
thus there are: 0 obsns not in the panel
covariance matrix :
0.11181666E+00 -0.34499581E-01 -0.49718410E-01 -0.16926567E-02 0.55024420E-02
0.91872486E-02 0.13903905E-02 0.21847833E-02
-0.34499581E-01 0.31833073E-01 0.10974810E-01 -0.28269643E-02 -0.42979281E-03
-0.69400207E-02 0.50854428E-03 0.19131869E-02
-0.49718410E-01 0.10974810E-01 0.33824460E-01 0.52094292E-03 -0.50665777E-02
-0.32011870E-02 0.92843854E-03 0.33773460E-02
-0.16926567E-02 -0.28269643E-02 0.52094292E-03 0.17541282E-02 -0.89314884E-04
0.11101174E-03 0.13032461E-03 0.48560825E-03
0.55024420E-02 -0.42979281E-03 -0.50665777E-02 -0.89314884E-04 0.88973655E-03
0.16360475E-03 -0.10084036E-03 -0.36404788E-03
0.91872486E-02 -0.69400207E-02 -0.32011870E-02 0.11101174E-03 0.16360475E-03
0.18097220E-02 -0.20444443E-03 -0.77172590E-03
0.13903905E-02 0.50854428E-03 0.92843854E-03 0.13032461E-03 -0.10084036E-03
-0.20444443E-03 0.32248641E-02 0.67679010E-02
0.21847833E-02 0.19131869E-02 0.33773460E-02 0.48560825E-03 -0.36404788E-03
-0.77172590E-03 0.67679010E-02 0.23274731E-01
technical efficiency estimates :
firm eff.-est.
1 0.74084312E+00
2 0.82240262E+00
3 0.72171467E+00
4 0.76681270E+00
5 0.77082303E+00
6 0.75598615E+00
7 0.71018644E+00
8 0.75355893E+00
9 0.83296448E+00
10 0.74089161E+00
11 0.55327831E+00
12 0.93494151E+00
13 0.49398394E+00
14 0.69063898E+00
15 0.91081565E+00
16 0.53370691E+00
17 0.76275093E+00
18 0.73407210E+00
19 0.83874142E+00
20 0.80338585E+00
21 0.68975204E+00
22 0.86518583E+00
23 0.80813678E+00
24 0.82172554E+00
25 0.64203544E+00
26 0.88132676E+00
27 0.82536749E+00
28 0.78521011E+00
29 0.85050138E+00
30 0.63889901E+00
31 0.61389880E+00
32 0.76555506E+00
33 0.87215034E+00
34 0.46934763E+00
35 0.35428298E+00
36 0.88612182E+00
37 0.84919407E+00
38 0.74839411E+00
39 0.65232559E+00
40 0.86660888E+00
41 0.80846384E+00
42 0.74213675E+00
43 0.79650388E+00
44 0.90259023E+00
45 0.72247961E+00
46 0.74079757E+00
47 0.86495277E+00
48 0.84967177E+00
49 0.68703129E+00
50 0.59855415E+00
51 0.82571375E+00
52 0.87618415E+00
53 0.87871638E+00
54 0.75229300E+00
55 0.78793269E+00
56 0.78020988E+00
57 0.85118217E+00
58 0.72445766E+00
59 0.87915699E+00
60 0.73573609E+00
mean efficiency = 0.75938806E+00
京公网安备 11010802022788号
