パネルデータの分析(アウトプット）
Source | SS df MS Number of obs = 315
-------------+------------------------------ F( 1, 313) = 584.04
Model | 372.899728 1 372.899728 Prob > F = 0.0000
Residual | 199.844534 313 .63848094 R-squared = 0.6511
-------------+------------------------------ Adj R-squared = 0.6500
Total | 572.744262 314 1.82402631 Root MSE = .79905
------------------------------------------------------------------------------
lp | Coef. Std. Err. t P>|t| [95% Conf. Interval]
-------------+----------------------------------------------------------------
rnd | .8454434 .0349834 24.17 0.000 .776611 .9142759
_cons | 1.438115 .1026224 14.01 0.000 1.236198 1.640032
------------------------------------------------------------------------------
. xtreg lp rnd, be i(cusip) ;
Between regression (regression on group means) Number of obs = 315
Group variable (i) : cusip Number of groups = 45
R-sq: within = 0.0208 Obs per group: min = 7
between = 0.7109 avg = 7.0
overall = 0.6511 max = 7
F(1,43) = 105.76
sd(u_i + avg(e_i.))= .7173364 Prob > F = 0.0000
------------------------------------------------------------------------------
lp | Coef. Std. Err. t P>|t| [95% Conf. Interval]
-------------+----------------------------------------------------------------
rnd | .871428 .084737 10.28 0.000 .7005396 1.042316
_cons | 1.369617 .247651 5.53 0.000 .8701816 1.869053
------------------------------------------------------------------------------
. xtreg lp rnd, fe i(cusip) ;
Fixed-effects (within) regression Number of obs = 315
Group variable (i) : cusip Number of groups = 45
R-sq: within = 0.0208 Obs per group: min = 7
between = 0.7109 avg = 7.0
overall = 0.6511 max = 7
F(1,269) = 5.71
corr(u_i, Xb) = 0.7575 Prob > F = 0.0176
------------------------------------------------------------------------------
lp | Coef. Std. Err. t P>|t| [95% Conf. Interval]
-------------+----------------------------------------------------------------
rnd | .1955142 .0818544 2.39 0.018 .0343574 .3566711
_cons | 3.151387 .2167619 14.54 0.000 2.724621 3.578152
-------------+----------------------------------------------------------------
sigma_u | 1.1166797
sigma_e | .36657743
rho | .90271926 (fraction of variance due to u_i)
------------------------------------------------------------------------------
F test that all u_i=0: F(44, 269) = 27.69 Prob > F = 0.0000
. xtreg lp rnd, re i(cusip) ;
Random-effects GLS regression Number of obs = 315
Group variable (i) : cusip Number of groups = 45
R-sq: within = 0.0208 Obs per group: min = 7
between = 0.7109 avg = 7.0
overall = 0.6511 max = 7
Random effects u_i ~ Gaussian Wald chi2(1) = 71.28
corr(u_i, X) = 0 (assumed) Prob > chi2 = 0.0000
------------------------------------------------------------------------------
lp | Coef. Std. Err. z P>|z| [95% Conf. Interval]
-------------+----------------------------------------------------------------
rnd | .5217792 .061801 8.44 0.000 .4006515 .642907
_cons | 2.291323 .1978421 11.58 0.000 1.903559 2.679086
-------------+----------------------------------------------------------------
sigma_u | .70382842
sigma_e | .36657743
rho | .78661648 (fraction of variance due to u_i)
------------------------------------------------------------------------------
. xthausman;
Hausman specification test
---- Coefficients ----
| Fixed Random
lp | Effects Effects Difference
-------------+-----------------------------------------
rnd | .1955142 .5217792 -.326265
Test: Ho: difference in coefficients not systematic
chi2( 1) = (b-B)'[S^(-1)](b-B), S = (S_fe - S_re)
= 36.95
Prob>chi2 = 0.0000