- 阅读权限
- 255
- 威望
- 0 级
- 论坛币
- 2822 个
- 通用积分
- 8.5001
- 学术水平
- 6 点
- 热心指数
- 5 点
- 信用等级
- 1 点
- 经验
- 3359 点
- 帖子
- 224
- 精华
- 0
- 在线时间
- 142 小时
- 注册时间
- 2005-8-24
- 最后登录
- 2024-12-26
已卖:447份资源
博士生
还不是VIP/贵宾
- 威望
- 0 级
- 论坛币
 - 2822 个
- 通用积分
- 8.5001
- 学术水平
- 6 点
- 热心指数
- 5 点
- 信用等级
- 1 点
- 经验
- 3359 点
- 帖子
- 224
- 精华
- 0
- 在线时间
- 142 小时
- 注册时间
- 2005-8-24
- 最后登录
- 2024-12-26
 | 开心 2018-4-24 17:16:38 |
|---|
签到天数: 8 天 连续签到: 1 天 [LV.3]偶尔看看II
|
3论坛币
|
各位大虾:
因数据是panel data,故需判断是用随机效应还是固定效应模型。
当检验结果做出来后,我看不懂。
麻烦大家看数据后,告诉如下两个问题答案。
1. hausman test 值是多少?根据这个数据,该采用什么模型?
2.为什么选择这个模型?
在线等,谢谢!
数据如下:
xtreg longunemrate eplfreetoworld loggdp perddp if year>2004,fe
Fixed-effects (within) regression Number of obs = 94
Group variable: year Number of groups = 3
R-sq: within = 0.2006 Obs per group: min = 31
between = 0.6115 avg = 31.3
overall = 0.2016 max = 32
F(3,88) = 7.36
corr(u_i, Xb) = -0.0779 Prob > F = 0.0002
------------------------------------------------------------------------------
longunemrate | Coef. Std. Err. t P>|t| [95% Conf. Interval]
-------------+----------------------------------------------------------------
eplfreetow~d | .3141716 .1417594 2.22 0.029 .0324546 .5958885
loggdp | -2.481883 1.219291 -2.04 0.045 -4.904968 -.0587981
perddp | -.0002583 .000097 -2.66 0.009 -.000451 -.0000655
_cons | 95.49042 33.84686 2.82 0.006 28.2269 162.7539
-------------+----------------------------------------------------------------
sigma_u | .98842057
sigma_e | 17.076294
rho | .00333921 (fraction of variance due to u_i)
------------------------------------------------------------------------------
F test that all u_i=0: F(2, 88) = 0.10 Prob > F = 0.9022
. est store fe
. xtreg longunemrate eplfreetoworld loggdp perddp if year>2004,re
Random-effects GLS regression Number of obs = 94
Group variable: year Number of groups = 3
R-sq: within = 0.2006 Obs per group: min = 31
between = 0.6136 avg = 31.3
overall = 0.2016 max = 32
Random effects u_i ~ Gaussian Wald chi2(3) = 22.73
corr(u_i, X) = 0 (assumed) Prob > chi2 = 0.0000
------------------------------------------------------------------------------
longunemrate | Coef. Std. Err. z P>|z| [95% Conf. Interval]
-------------+----------------------------------------------------------------
eplfreetow~d | .3094663 .139949 2.21 0.027 .0351713 .5837612
loggdp | -2.469466 1.204047 -2.05 0.040 -4.829355 -.1095767
perddp | -.0002576 .000095 -2.71 0.007 -.0004439 -.0000713
_cons | 95.32785 33.39519 2.85 0.004 29.87448 160.7812
-------------+----------------------------------------------------------------
sigma_u | 0
sigma_e | 17.076294
rho | 0 (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.
-------------+----------------------------------------------------------------
eplfreetow~d | .3141716 .3094663 .0047053 .0225833
loggdp | -2.481883 -2.469466 -.0124175 .192202
perddp | -.0002583 -.0002576 -6.70e-07 .0000194
------------------------------------------------------------------------------
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(3) = (b-B)'[(V_b-V_B)^(-1)](b-B)
= 0.05
Prob>chi2 = 0.9969 |
|