利用huausman检验面板数据内生性是否存在时,结果为负且没有显示p值
希望各位老师指点,谢谢你们! 以下是stata结果。。。。
. xtreg TL1 lnbrand wppatent LNPGDP open2 soe , fe
Fixed-effects (within) regression Number of obs = 465
Group variable: provi Number of groups = 31
R-sq: Obs per group:
within = 0.3056 min = 15
between = 0.2576 avg = 15.0
overall = 0.0279 max = 15
F(5,429) = 37.77
corr(u_i, Xb) = -0.4274 Prob > F = 0.0000
------------------------------------------------------------------------------
TL1 | Coef. Std. Err. t P>|t| [95% Conf. Interval]
-------------+----------------------------------------------------------------
lnbrand | .6173789 .1948867 3.17 0.002 .2343274 1.00043
wppatent | .2208509 .032952 6.70 0.000 .1560834 .2856183
LNPGDP | -3.373785 1.07699 -3.13 0.002 -5.490618 -1.256953
open2 | -3.000428 1.472878 -2.04 0.042 -5.895384 -.1054723
soe | -.2478109 .1074765 -2.31 0.022 -.4590568 -.0365649
_cons | 35.87641 9.498558 3.78 0.000 17.20691 54.54592
-------------+----------------------------------------------------------------
sigma_u | 8.5590016
sigma_e | 2.5999525
rho | .91552026 (fraction of variance due to u_i)
------------------------------------------------------------------------------
F test that all u_i=0: F(30, 429) = 33.03 Prob > F = 0.0000
.
. estimates store fe
.
. xtivreg TL1 wppatent LNPGDP open2 soe (lnbrand=L.lnbrand), fe
Fixed-effects (within) IV regression Number of obs = 434
Group variable: provi Number of groups = 31
R-sq: Obs per group:
within = 0.3134 min = 14
between = 0.3047 avg = 14.0
overall = 0.0923 max = 14
Wald chi2(5) = 3767.71
corr(u_i, Xb) = -0.5788 Prob > chi2 = 0.0000
------------------------------------------------------------------------------
TL1 | Coef. Std. Err. z P>|z| [95% Conf. Interval]
-------------+----------------------------------------------------------------
lnbrand | 1.04855 .3163032 3.32 0.001 .4286071 1.668493
wppatent | .2121652 .0371197 5.72 0.000 .1394119 .2849185
LNPGDP | -5.912044 1.285789 -4.60 0.000 -8.432144 -3.391945
open2 | -2.304316 1.533355 -1.50 0.133 -5.309636 .7010043
soe | -.5301802 .196464 -2.70 0.007 -.9152425 -.1451179
_cons | 58.61467 11.37075 5.15 0.000 36.32841 80.90094
-------------+----------------------------------------------------------------
sigma_u | 9.3289086
sigma_e | 2.4600183
rho | .93498415 (fraction of variance due to u_i)
------------------------------------------------------------------------------
F test that all u_i=0: F(30,398) = 36.05 Prob > F = 0.0000
------------------------------------------------------------------------------
Instrumented: lnbrand
Instruments: wppatent LNPGDP open2 soe L.lnbrand
------------------------------------------------------------------------------
.
. estimates store fe_iv
.
. hausman fe_iv fe
---- Coefficients ----
| (b) (B) (b-B) sqrt(diag(V_b-V_B))
| fe_iv fe Difference S.E.
-------------+----------------------------------------------------------------
lnbrand | 1.04855 .6173789 .4311711 .2491323
wppatent | .2121652 .2208509 -.0086857 .0170891
LNPGDP | -5.912044 -3.373785 -2.538259 .7023861
open2 | -2.304316 -3.000428 .6961122 .4263876
soe | -.5301802 -.2478109 -.2823693 .1644594
------------------------------------------------------------------------------
b = consistent under Ho and Ha; obtained from xtivreg
B = inconsistent under Ha, efficient under Ho; obtained from xtreg
Test: Ho: difference in coefficients not systematic
chi2(5) = (b-B)'[(V_b-V_B)^(-1)](b-B)
= -30.92 chi2<0 ==> model fitted on these
data fails to meet the asymptotic
assumptions of the Hausman test;
see suest for a generalized test


雷达卡








京公网安备 11010802022788号







