黃河泉 发表于 2019-3-23 10:41 
1. 的确是不太一样。3. 你用一般之 hausman test 之结果可否 show 一下? 3. 你的 CF1 是啥？
这是现金-现金流敏感性模型，被解释变量为cash,是现金及现金等价物增加额，，cf是经营活动现金流量净额，cf1是对cf进行中心化处理之后的值，命令为center cf,gen(cf1)
一般的hausman test仅适用于普通标准误的固定效应和随机效应检验，老师，我不采用稳健标准误可以吗？而且一般的hausman test结果还出现一个note，是说我的结果有问题吧。下面随机效应回归结果 sigma_u=0是不是说我不存在随机效应，得用混合回归啊？但是我混合回归没有F和P值。
一般的hausman test 命令如下：
xtreg cash cf1 var1 cscore income invest debt tobinq,fe
est store FE
xtreg cash cf1 var1 cscore income invest debt tobinq,re
est store RE
hausman FE RE,constant sigmamore
结果如下：
. xtreg cash cf1 var1 cscore income invest debt tobinq,fe
Fixed-effects (within) regression Number of obs = 347
Group variable: fcode Number of groups = 39
R-sq: within = 0.1411 Obs per group: min = 2
between = 0.0228 avg = 8.9
overall = 0.0838 max = 11
F(7,301) = 7.06
corr(u_i, Xb) = -0.5209 Prob > F = 0.0000
------------------------------------------------------------------------------
cash | Coef. Std. Err. t P>|t| [95% Conf. Interval]
-------------+----------------------------------------------------------------
cf1 | .2377988 .1054167 2.26 0.025 .0303517 .4452459
var1 | 2.89622 1.344419 2.15 0.032 .2505699 5.54187
cscore | -.4738704 .1867491 -2.54 0.012 -.8413696 -.1063712
income | .0991569 .0337253 2.94 0.004 .0327897 .1655241
invest | -.5373448 .150439 -3.57 0.000 -.8333903 -.2412994
debt | .2749575 .0841317 3.27 0.001 .1093967 .4405183
tobinq | .00792 .0054044 1.47 0.144 -.0027151 .0185551
_cons | .0134024 .020074 0.67 0.505 -.0261009 .0529056
-------------+----------------------------------------------------------------
sigma_u | .04590076
sigma_e | .11581287
rho | .13575694 (fraction of variance due to u_i)
------------------------------------------------------------------------------
F test that all u_i=0: F(38, 301) = 0.61 Prob > F = 0.9660
. est store FE
. xtreg cash cf1 var1 cscore income invest debt tobinq,re
Random-effects GLS regression Number of obs = 347
Group variable: fcode Number of groups = 39
R-sq: within = 0.1215 Obs per group: min = 2
between = 0.0081 avg = 8.9
overall = 0.1044 max = 11
Wald chi2(7) = 39.51
corr(u_i, X) = 0 (assumed) Prob > chi2 = 0.0000
------------------------------------------------------------------------------
cash | Coef. Std. Err. z P>|z| [95% Conf. Interval]
-------------+----------------------------------------------------------------
cf1 | .2007698 .0819267 2.45 0.014 .0401964 .3613432
var1 | 2.606168 .9441395 2.76 0.006 .7556885 4.456647
cscore | -.1190467 .1111854 -1.07 0.284 -.3369661 .0988727
income | .1017463 .0307477 3.31 0.001 .0414819 .1620108
invest | -.2551542 .1059695 -2.41 0.016 -.4628506 -.0474579
debt | .2380131 .0786103 3.03 0.002 .0839397 .3920866
tobinq | .0057896 .0041829 1.38 0.166 -.0024086 .0139878
_cons | -.0024139 .0157918 -0.15 0.879 -.0333652 .0285373
-------------+----------------------------------------------------------------
sigma_u | 0
sigma_e | .11581287
rho | 0 (fraction of variance due to u_i)
------------------------------------------------------------------------------
. est store RE
. hausman FE RE,constant sigmamore
Note: the rank of the differenced variance matrix (7) does not equal the number of coefficients being tested (8); 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.
-------------+----------------------------------------------------------------
cf1 | .2377988 .2007698 .037029 .0625901
var1 | 2.89622 2.606168 .2900518 .9151354
cscore | -.4738704 -.1190467 -.3548237 .1449022
income | .0991569 .1017463 -.0025894 .0119381
invest | -.5373448 -.2551542 -.2821906 .1020702
debt | .2749575 .2380131 .0369444 .0243063
tobinq | .00792 .0057896 .0021304 .0032312
_cons | .0134024 -.0024139 .0158163 .0116648
------------------------------------------------------------------------------
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(7) = (b-B)'[(V_b-V_B)^(-1)](b-B)
= 15.30
Prob>chi2 = 0.0324
(V_b-V_B is not positive definite)
.
end of do-file
混合回归结果如下：
reg cash cf1 income invest debt tobinq i.fcode,vce(cluster fcode)
Linear regression Number of obs = 347
F( 4, 38) = .
Prob > F = .
R-squared = 0.1370
Root MSE = .1176
(Std. Err. adjusted for 39 clusters in fcode)
------------------------------------------------------------------------------
| Robust
cash | Coef. Std. Err. t P>|t| [95% Conf. Interval]
-------------+----------------------------------------------------------------
cf1 | .2748435 .0872558 3.15 0.003 .0982035 .4514835
income | .1106586 .0447857 2.47 0.018 .0199947 .2013225
invest | -.4979956 .1391322 -3.58 0.001 -.779654 -.2163371
debt | .2531551 .0876024 2.89 0.006 .0758132 .430497
tobinq | .0125237 .0066106 1.89 0.066 -.0008588 .0259061
|
fcode |
713 | -.0141136 .011187 -1.26 0.215 -.0367605 .0085333
735 | .0201562 .012859 1.57 0.125 -.0058755 .046188
798 | -.0559216 .0069849 -8.01 0.000 -.0700617 -.0417815
998 | -.0126873 .0080206 -1.58 0.122 -.0289242 .0035495
2041 | -.0209428 .0124646 -1.68 0.101 -.046176 .0042905
2069 | -.0290916 .0048893 -5.95 0.000 -.0389894 -.0191938
2086 | .0171118 .0139415 1.23 0.227 -.0111112 .0453349
2200 | -.0082108 .0052188 -1.57 0.124 -.0187757 .0023541
2234 | .0151796 .0139296 1.09 0.283 -.0130194 .0433786
2299 | .0351601 .0208876 1.68 0.101 -.0071247 .0774449
2321 | .0348431 .0185695 1.88 0.068 -.002749 .0724352
2458 | .0001392 .0113741 0.01 0.990 -.0228864 .0231647
2477 | .0527061 .0144879 3.64 0.001 .0233769 .0820354
2679 | .0215856 .0087982 2.45 0.019 .0037745 .0393967
2696 | .0125176 .0152365 0.82 0.416 -.018327 .0433622
2714 | .101039 .0300434 3.36 0.002 .0402192 .1618587
2746 | -.0698255 .019435 -3.59 0.001 -.1091696 -.0304813
2772 | .0760087 .0151497 5.02 0.000 .0453396 .1066777
300087 | .0350502 .0073098 4.79 0.000 .0202524 .0498481
300094 | -.0055966 .0114693 -0.49 0.628 -.0288151 .0176219
300106 | .0166951 .0111684 1.49 0.143 -.0059141 .0393043
300189 | .0396689 .0134085 2.96 0.005 .0125247 .0668131
300498 | -.1177091 .0306696 -3.84 0.000 -.1797964 -.0556217
300511 | .053999 .0336596 1.60 0.117 -.0141412 .1221393
600097 | .002325 .0133311 0.17 0.862 -.0246625 .0293125
600108 | -.0036767 .0099159 -0.37 0.713 -.0237504 .016397
600257 | -.0036215 .0059437 -0.61 0.546 -.015654 .008411
600313 | -.0372179 .0114589 -3.25 0.002 -.0604151 -.0140206
600354 | .0042925 .0147613 0.29 0.773 -.0255903 .0341753
600359 | -.0055057 .012898 -0.43 0.672 -.0316163 .0206049
600371 | -.0117425 .0092677 -1.27 0.213 -.030504 .007019
600467 | .0016387 .0126736 0.13 0.898 -.0240176 .0272949
600506 | -.0752079 .0179989 -4.18 0.000 -.1116449 -.038771
600540 | -.0019961 .0138773 -0.14 0.886 -.0300892 .0260969
600598 | -.0235021 .0136335 -1.72 0.093 -.0511017 .0040975
600965 | -.0400433 .0064927 -6.17 0.000 -.053187 -.0268996
600975 | -.0340634 .0070649 -4.82 0.000 -.0483655 -.0197612
601118 | -.0464356 .0090965 -5.10 0.000 -.0648505 -.0280207
|
_cons | -.0116483 .0258806 -0.45 0.655 -.0640409 .0407443
------------------------------------------------------------------------------
.
end of do-file
扫码加好友，拉您进群
全部版块
我的主页
收藏
