固定效应:
note: lnD omitted because of collinearity
note: F omitted because of collinearity
note: L omitted because of collinearity
Fixed-effects (within) regression Number of obs = 680
Group variable: codeId Number of groups = 60
R-sq: within = 0.7392 Obs per group: min = 1
between = 0.0036 avg = 11.3
overall = 0.0245 max = 13
F(9,611) = 192.38
corr(u_i, Xb) = -0.6837 Prob > F = 0.0000
------------------------------------------------------------------------------
lnX | Coef. Std. Err. t P>|t| [95% Conf. Interval]
-------------+----------------------------------------------------------------
lngdpi | 5.685667 .5901933 9.63 0.000 4.526613 6.84472
lngdpj | 1.130826 .2116431 5.34 0.000 .7151895 1.546462
lnpopi | -65.82947 9.854426 -6.68 0.000 -85.18213 -46.47682
lnpopj | -1.63651 .4524184 -3.62 0.000 -2.524994 -.748026
lnD | 0 (omitted)
F | 0 (omitted)
L | 0 (omitted)
T | -.0180466 .0086565 -2.08 0.038 -.0350467 -.0010465
lnHC | -4.606837 .8648479 -5.33 0.000 -6.305272 -2.908402
lnLT | .2397727 .2047864 1.17 0.242 -.162398 .6419434
lnIPR | 2.780269 .6924933 4.01 0.000 1.420313 4.140225
lnLTIPR | -.2497586 .0515337 -4.85 0.000 -.3509632 -.1485539
_cons | 1245.262 197.4892 6.31 0.000 857.4222 1633.102
-------------+----------------------------------------------------------------
sigma_u | 2.569473
sigma_e | .32296767
rho | .98444671 (fraction of variance due to u_i)
------------------------------------------------------------------------------
F test that all u_i=0: F(59, 611) = 120.70 Prob > F = 0.0000
.
随机效应:
Random-effects GLS regression Number of obs = 680
Group variable: codeId Number of groups = 60
R-sq: within = 0.7323 Obs per group: min = 1
between = 0.8320 avg = 11.3
overall = 0.8138 max = 13
Wald chi2(12) = 1940.59
corr(u_i, X) = 0 (assumed) Prob > chi2 = 0.0000
------------------------------------------------------------------------------
lnX | Coef. Std. Err. z P>|z| [95% Conf. Interval]
-------------+----------------------------------------------------------------
lngdpi | 5.833693 .5873254 9.93 0.000 4.682556 6.98483
lngdpj | .8930958 .0889653 10.04 0.000 .7187269 1.067465
lnpopi | -66.00707 9.944787 -6.64 0.000 -85.49849 -46.51565
lnpopj | .0983778 .1004435 0.98 0.327 -.0984879 .2952434
lnD | -.5214865 .1909556 -2.73 0.006 -.8957526 -.1472204
F | .2883289 .3896035 0.74 0.459 -.4752798 1.051938
L | 2.754753 .4885587 5.64 0.000 1.797195 3.71231
T | -.0185099 .0082402 -2.25 0.025 -.0346604 -.0023594
lnHC | -4.867938 .8690074 -5.60 0.000 -6.571161 -3.164715
lnLT | .1818767 .2050249 0.89 0.375 -.2199647 .5837182
lnIPR | 3.070071 .6255919 4.91 0.000 1.843933 4.296208
lnLTIPR | -.2625025 .047134 -5.57 0.000 -.3548836 -.1701215
_cons | 1226.877 199.1998 6.16 0.000 836.4523 1617.301
-------------+----------------------------------------------------------------
sigma_u | .76385811
sigma_e | .32296767
rho | .84834249 (fraction of variance due to u_i)
------------------------------------------------------------------------------
.
Hausman检验:
Note: the rank of the differenced variance matrix (8) does not equal the number of coefficients
being tested (10); be sure this is what you expect, or there may be problems computing the
test. Examine the output of your estimators for anything unexpected and possibly consider
scaling your variables so that the coefficients are on a similar scale.
---- Coefficients ----
| (b) (B) (b-B) sqrt(diag(V_b-V_B))
| fe re Difference S.E.
-------------+----------------------------------------------------------------
lngdpi | 5.685667 5.833693 -.1480262 .1020798
lngdpj | 1.130826 .8930958 .2377298 .1943803
lnpopi | -65.82947 -66.00707 .1775967 .4177836
lnpopj | -1.63651 .0983778 -1.734888 .445794
T | -.0180466 -.0185099 .0004632 .0029239
lnHC | -4.606837 -4.867938 .2611009 .088951
lnLT | .2397727 .1818767 .057896 .027391
lnIPR | 2.780269 3.070071 -.2898014 .3128554
lnLTIPR | -.2497586 -.2625025 .012744 .0220862
_cons | 1245.262 1226.877 18.38542 10.49137
------------------------------------------------------------------------------
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(8) = (b-B)'[(V_b-V_B)^(-1)](b-B)
= 15.55
Prob>chi2 = 0.0492
(V_b-V_B is not positive definite)
.
根据Hausman检验p小于0.05,拒绝随机效应,使用固定效应。但是我的固定效应有三个参数被忽略了,但是我需要这几个参数都存在而不是被忽略,所以比较纠结,各位大神能不能给解释一下怎么弄?stata新手太菜。


雷达卡





京公网安备 11010802022788号







