我作出的检验结果如下:
xi: xtreg bf GDP ES pop urb inv inc i.yr if district == 1, fe
i.yr _Iyr_1999-2007 (naturally coded; _Iyr_1999 omitted)
Fixed-effects (within) regression Number of obs = 117
Group variable: id Number of groups = 13
R-sq: within = 0.9512 Obs per group: min = 9
between = 0.8240 avg = 9.0
overall = 0.8097 max = 9
F(14,90) = 125.35
corr(u_i, Xb) = 0.4810 Prob > F = 0.0000
------------------------------------------------------------------------------
bf | Coef. Std. Err. t P>|t| [95% Conf. Interval]
-------------+----------------------------------------------------------------
GDP | .2161294 .1852852 1.17 0.247 -.1519721 .5842308
ES | -.5959407 .7742343 -0.77 0.443 -2.134092 .9422109
pop | .4243134 .460748 0.92 0.360 -.4910429 1.33967
urb | .1930201 .1186234 1.63 0.107 -.042646 .4286862
inv | .0042507 .0264612 0.16 0.873 -.048319 .0568205
inc | .4681348 .3309421 1.41 0.161 -.1893395 1.125609
_Iyr_2000 | .0100367 .052526 0.19 0.849 -.0943153 .1143887
_Iyr_2001 | .0938836 .0712513 1.32 0.191 -.0476695 .2354368
_Iyr_2002 | .1148888 .1005915 1.14 0.256 -.0849537 .3147314
_Iyr_2003 | .1423608 .134488 1.06 0.293 -.1248231 .4095448
_Iyr_2004 | .1874946 .174244 1.08 0.285 -.1586716 .5336607
_Iyr_2005 | .2284545 .2136036 1.07 0.288 -.1959063 .6528152
_Iyr_2006 | .3426005 .2550774 1.34 0.183 -.1641552 .8493562
_Iyr_2007 | .5617449 .2997844 1.87 0.064 -.0338292 1.157319
_cons | -2.497945 4.919014 -0.51 0.613 -12.27043 7.274535
-------------+----------------------------------------------------------------
sigma_u | .45845055
sigma_e | .10110265
rho | .95362156 (fraction of variance due to u_i)
------------------------------------------------------------------------------
F test that all u_i=0: F(12, 90) = 13.52 Prob > F = 0.0000
. xi: xtreg bf GDP ES pop urb inv inc i.yr if district == 1, re
i.yr _Iyr_1999-2007 (naturally coded; _Iyr_1999 omitted)
Random-effects GLS regression Number of obs = 117
Group variable: id Number of groups = 13
R-sq: within = 0.9447 Obs per group: min = 9
between = 0.9782 avg = 9.0
overall = 0.9713 max = 9
Random effects u_i ~ Gaussian Wald chi2(14) = 2043.82
corr(u_i, X) = 0 (assumed) Prob > chi2 = 0.0000
------------------------------------------------------------------------------
bf | Coef. Std. Err. z P>|z| [95% Conf. Interval]
-------------+----------------------------------------------------------------
GDP | .44401 .1604448 2.77 0.006 .129544 .7584759
ES | .2585905 .5488239 0.47 0.638 -.8170847 1.334266
pop | .3909278 .1578337 2.48 0.013 .0815794 .7002761
urb | .269244 .0985091 2.73 0.006 .0761697 .4623183
inv | .0092081 .0271231 0.34 0.734 -.0439521 .0623684
inc | 1.178017 .2023247 5.82 0.000 .781468 1.574566
_Iyr_2000 | -.078156 .0433988 -1.80 0.072 -.1632161 .006904
_Iyr_2001 | -.0883334 .0478718 -1.85 0.065 -.1821604 .0054936
_Iyr_2002 | -.1679904 .0568507 -2.95 0.003 -.2794157 -.056565
_Iyr_2003 | -.2561843 .0678278 -3.78 0.000 -.3891243 -.1232444
_Iyr_2004 | -.3351011 .0869018 -3.86 0.000 -.5054255 -.1647768
_Iyr_2005 | -.424235 .1018657 -4.16 0.000 -.6238881 -.2245818
_Iyr_2006 | -.4401889 .1190665 -3.70 0.000 -.673555 -.2068227
_Iyr_2007 | -.3604594 .1373498 -2.62 0.009 -.6296601 -.0912586
_cons | -10.47976 2.069803 -5.06 0.000 -14.5365 -6.423025
-------------+----------------------------------------------------------------
sigma_u | .11637788
sigma_e | .10110265
rho | .56989242 (fraction of variance due to u_i)
------------------------------------------------------------------------------
. est store re
. hausman fe
---- Coefficients ----
| (b) (B) (b-B) sqrt(diag(V_b-V_B))
| fe re Difference S.E.
-------------+----------------------------------------------------------------
GDP | .2161294 .44401 -.2278806 .092672
ES | -.5959407 .2585905 -.8545312 .5461054
pop | .4243134 .3909278 .0333857 .4328709
urb | .1930201 .269244 -.0762239 .0660868
inv | .0042507 .0092081 -.0049574 .
inc | .4681348 1.178017 -.7098823 .261892
_Iyr_2000 | .0100367 -.078156 .0881927 .0295892
_Iyr_2001 | .0938836 -.0883334 .1822171 .0527735
_Iyr_2002 | .1148888 -.1679904 .2828792 .0829858
_Iyr_2003 | .1423608 -.2561843 .3985452 .116131
_Iyr_2004 | .1874946 -.3351011 .5225957 .1510267
_Iyr_2005 | .2284545 -.424235 .6526894 .1877495
_Iyr_2006 | .3426005 -.4401889 .7827894 .2255828
_Iyr_2007 | .5617449 -.3604594 .9222043 .266469
------------------------------------------------------------------------------
b = consistent under Ho and Ha; obtained from xtreg
B = inconsistent under Ha, efficient under Ho; obtained from xtreg
Test: Ho: difference in coefficients not systematic
chi2(14) = (b-B)'[(V_b-V_B)^(-1)](b-B)
= 13.62
Prob>chi2 = 0.4785
(V_b-V_B is not positive definite)
我看了论坛上很多解释,好像选择FE还是re意见不太统一,有高人能指点一下吗?


雷达卡



京公网安备 11010802022788号







