hausman test, 请教各位高手,很快要交论文了!!

是 否 +2 论坛币

k人 参与回答

经管之家送您一份

应届毕业生专属福利!

求职就业群

赵安豆老师微信：zhaoandou666

经管之家联合CDA

送您一个全额奖学金名额~ !

立即领取

感谢您参与论坛问题回答

经管之家送您两个论坛币！

+2 论坛币

<P>我用EVIEW hausman 检验结果如下,请教高手检验结果是什么意思啊,应该用什么模型啊???</P>
<P>Correlated Random Effects - Hausman Test    <BR>Pool: GROWTH    <BR>Test cross-section random effects    <BR>    <BR>Test Summary  Chi-Sq. Statistic Chi-Sq. d.f. Prob. <BR>    <BR>Cross-section random  0.000000 5 1.0000<BR>    <BR>* Cross-section test variance is invalid. Hausman statistic set to zero.    <BR>** Warning: robust standard errors may not be consistent with    <BR>        assumptions of Hausman test variance calculation.    <BR>    <BR>Cross-section random effects test comparisons:    <BR>    <BR>Variable Fixed            Random        Var(Diff.)     Prob. <BR>    <BR>FDI?   0.537031       0.615760      -0.004625      NA<BR>H?    47.733719         72.613749   -46.411392   NA<BR>FDIH? -144.371892   -316.654325 -5389.533238 NA<BR>GC? -1.094572          -0.497975      -0.032002    NA<BR>Y0? 1.025828            1.034355       -0.000008    NA<BR>    <BR>    <BR>Cross-section random effects test equation:    <BR>Dependent Variable: LNY?    <BR>Method: Panel Least Squares    <BR>Date: 03/15/07   Time: 20:58    <BR>Sample: 1993 2003    <BR>Included observations: 11    <BR>Cross-sections included: 29    <BR>Total pool (balanced) observations: 319    <BR>Cross-section SUR (PCSE) standard errors & covariance (d.f.    <BR>        corrected)    <BR>    <BR>Variable   Coefficient    Std. Error     t-Statistic    Prob.  <BR>    <BR>C              0.351022   0.186342   1.883750      0.0607<BR>FDI?          0.537031   0.251821   2.132589    0.0338<BR>H?           47.73372      9.997561    4.774536   0.0000<BR>FDIH?     -144.3719     176.0432    -0.820094    0.4129<BR>GC?       -1.094572      0.240883     -4.544001    0.0000<BR>Y0?          1.025828      0.025636      40.01503    0.0000<BR>    <BR> Effects Specification   <BR>    <BR>Cross-section fixed (dummy variables)    <BR>Period fixed (dummy variables)    <BR>    <BR>R-squared 0.969127     Mean dependent var  8.069561<BR>Adjusted R-squared 0.964300     S.D. dependent var  0.595917<BR>S.E. of regression 0.112596     Akaike info criterion  -1.402585<BR>Sum squared resid 3.486389     Schwarz criterion  -0.883249<BR>Log likelihood 267.7124     F-statistic  200.7559<BR>Durbin-Watson stat 1.789136     Prob(F-statistic)  0.000000<BR>    <BR></P>

扫码加我 拉你入群

请注明：姓名-公司-职位

以便审核进群资格，未注明则拒绝

分享0 收藏0 回帖

关键词：hausman ausman test Man Aus standard invalid errors Random 论文

Can you read P value, then deciding whether or not you need to reject some lags ...

Use Eviews to test Panal Data....

Such as Hausman Test, Fixed Effect, Random Effect...

Correlated Random Effects - Hausman Test
Pool: GROWTH
Test cross-section random effects

Test Summary Chi-Sq. Statistic Chi-Sq.d.f. Prob.

Cross-section random 0.000000 5 1.0000

* Cross-section test variance is invalid. Hausman statistic set to zero.
** Warning: robust standard errors may not be consistent with
assumptions of Hausman test variance calculation.

请问这个结果什么意思啊?P=1? Chi-Sq=0? 后面这句话什么意思啊?

不能拒绝原假设，那就是用随机效应了。

你还不如用stata 做呢

EVIEWS的H检验，p值小于0.05，就可以认为是拒绝原假设，即选择个体固定效应模型~

当然，如果大于0.05的P值，可以认为是接受原假设，即用个体随机效应模型

http://dss.princeton.edu/training/Panel101.pdf 这个上面有详尽的解说

选择固体效应吧