我用frontier4.1估计技术效率，用的十年的数据，但是有个别年份的数据做出来的结果是相似的，比如都是0.9981、0.9983、0.9984之类的，这不符合实际情况啊，可是用相同的方法只有个别年份这样其他年份是正常的，这是怎么回事啊，向大神求教！这是我其中一年的结果
the ols estimates are :
coefficient standard-error t-ratio
beta 0 0.12823632E+01 0.22417339E+01 0.57204080E+00
beta 1 0.70501850E+00 0.15051598E+00 0.46840111E+01
beta 2 0.48545586E+00 0.12458782E+00 0.38964954E+01
beta 3 0.34029055E+00 0.81604253E-01 0.41700100E+01
sigma-squared 0.43516619E-01
log likelihood function = 0.67407283E+01
the estimates after the grid search were :
beta 0 0.13176635E+01
beta 1 0.70501850E+00
beta 2 0.48545586E+00
beta 3 0.34029055E+00
sigma-squared 0.39147681E-01
gamma 0.50000000E-01
mu is restricted to be zero
eta is restricted to be zero
iteration = 0 func evals = 20 llf = 0.67370323E+01
0.13176635E+01 0.70501850E+00 0.48545586E+00 0.34029055E+00 0.39147681E-01
0.50000000E-01
gradient step
iteration = 5 func evals = 49 llf = 0.67384830E+01
0.13176768E+01 0.70484256E+00 0.48520926E+00 0.34220865E+00 0.38661253E-01
0.30888919E-01
iteration = 10 func evals = 118 llf = 0.67403986E+01
0.12999831E+01 0.70494928E+00 0.48540060E+00 0.34083860E+00 0.38131703E-01
0.93579502E-02
iteration = 15 func evals = 222 llf = 0.67406428E+01
0.12868897E+01 0.70539739E+00 0.48512940E+00 0.34054645E+00 0.38001520E-01
0.39969653E-02
iteration = 20 func evals = 313 llf = 0.67407076E+01
0.12903791E+01 0.70489042E+00 0.48554833E+00 0.34035344E+00 0.37938038E-01
0.14628546E-02
iteration = 25 func evals = 420 llf = 0.67407219E+01
0.12872465E+01 0.70495837E+00 0.48550982E+00 0.34022667E+00 0.37916818E-01
0.66517326E-03
iteration = 30 func evals = 527 llf = 0.67407256E+01
0.12831969E+01 0.70515622E+00 0.48535424E+00 0.34025162E+00 0.37909626E-01
0.35594569E-03
iteration = 35 func evals = 638 llf = 0.67407275E+01
0.12832021E+01 0.70509029E+00 0.48540491E+00 0.34026920E+00 0.37905484E-01
0.16068745E-03
iteration = 40 func evals = 713 llf = 0.67407279E+01
0.12841757E+01 0.70500341E+00 0.48546807E+00 0.34028241E+00 0.37903966E-01
0.10331440E-03
iteration = 45 func evals = 807 llf = 0.67407281E+01
0.12833708E+01 0.70502668E+00 0.48545036E+00 0.34028246E+00 0.37902859E-01
0.53439224E-04
iteration = 48 func evals = 838 llf = 0.67407282E+01
0.12834190E+01 0.70501753E+00 0.48545660E+00 0.34029073E+00 0.37902644E-01
0.44506672E-04
the final mle estimates are :
coefficient standard-error t-ratio
beta 0 0.12834190E+01 0.19954391E+01 0.64317621E+00
beta 1 0.70501753E+00 0.13046806E+00 0.54037558E+01
beta 2 0.48545660E+00 0.10902742E+00 0.44526104E+01
beta 3 0.34029073E+00 0.74997831E-01 0.45373409E+01
sigma-squared 0.37902644E-01 0.10009501E-01 0.37866668E+01
gamma 0.44506672E-04 0.48423869E-01 0.91910607E-03
mu is restricted to be zero
eta is restricted to be zero
log likelihood function = 0.67407282E+01
the likelihood value is less than that obtained
using ols! - try again using different starting values
number of iterations = 48
(maximum number of iterations set at : 100)
number of cross-sections = 31
number of time periods = 1
total number of observations = 31
thus there are: 0 obsns not in the panel
covariance matrix :
0.39817771E+01 -0.24543566E+00 0.18509618E+00 0.26757445E-01 0.81626253E-03
0.23217304E-01
-0.24543566E+00 0.17021914E-01 -0.13673485E-01 -0.11638461E-02 -0.11958450E-04
0.16372867E-03
0.18509618E+00 -0.13673485E-01 0.11886977E-01 0.23034904E-04 0.86138293E-05
-0.12724883E-03
0.26757445E-01 -0.11638461E-02 0.23034904E-04 0.56246747E-02 -0.38139516E-05
0.22436091E-05
0.81626253E-03 -0.11958450E-04 0.86138293E-05 -0.38139516E-05 0.10019011E-03
0.57077880E-04
0.23217304E-01 0.16372867E-03 -0.12724883E-03 0.22436091E-05 0.57077880E-04
0.23448711E-02
technical efficiency estimates :
firm eff.-est.
1 0.99897049E+00
2 0.99896970E+00
3 0.99896286E+00
4 0.99896106E+00
5 0.99896022E+00
6 0.99896063E+00
7 0.99896344E+00
8 0.99895950E+00
9 0.99896915E+00
10 0.99896605E+00
11 0.99896540E+00
12 0.99896144E+00
13 0.99896362E+00
14 0.99896205E+00
15 0.99896361E+00
16 0.99896537E+00
17 0.99896215E+00
18 0.99896538E+00
19 0.99896463E+00
20 0.99896603E+00
21 0.99896981E+00
22 0.99896955E+00
23 0.99896354E+00
24 0.99896394E+00
25 0.99896811E+00
26 0.99895884E+00
27 0.99896255E+00
28 0.99896388E+00
29 0.99896379E+00
30 0.99896777E+00
31 0.99896614E+00
扫码加好友，拉您进群
全部版块
我的主页
收藏
