xtreg innov er1 er2 pro lngov lneco lnopen lncomp scale,fe
Fixed-effects (within) regression Number of obs = 359
Group variable: provi Number of groups = 30
R-sq: within = 0.0322 Obs per group: min = 11
between = 0.1425 avg = 12.0
overall = 0.0376 max = 12
F(8,321) = 1.34
corr(u_i, Xb) = -0.2576 Prob > F = 0.2251
------------------------------------------------------------------------------
innov | Coef. Std. Err. t P>|t| [95% Conf. Interval]
-------------+----------------------------------------------------------------
er1 | -1.989303 3.012137 -0.66 0.509 -7.915327 3.936721
er2 | -.0016415 .0193118 -0.08 0.932 -.0396352 .0363523
pro | -6.232694 13.03585 -0.48 0.633 -31.87919 19.4138
lngov | .1323202 .231778 0.57 0.568 -.3236755 .5883159
lneco | -.4286931 .5959604 -0.72 0.472 -1.601175 .7437885
lnopen | -.2145086 .0942608 -2.28 0.024 -.3999556 -.0290617
lncomp | .4562422 .3987591 1.14 0.253 -.3282693 1.240754
scale | .0222759 .0758596 0.29 0.769 -.126969 .1715208
_cons | 3.30663 4.301721 0.77 0.443 -5.156497 11.76976
-------------+----------------------------------------------------------------
sigma_u | .40798191
sigma_e | 1.3236944
rho | .08675484 (fraction of variance due to u_i)
------------------------------------------------------------------------------
F test that all u_i=0: F(29, 321) = 0.89 Prob > F = 0.6303
. xtreg innov er1 er2 pro lngov lneco lnopen lncomp scale,re
Random-effects GLS regression Number of obs = 359
Group variable: provi Number of groups = 30
R-sq: within = 0.0283 Obs per group: min = 11
between = 0.2987 avg = 12.0
overall = 0.0510 max = 12
Wald chi2(8) = 18.31
corr(u_i, X) = 0 (assumed) Prob > chi2 = 0.0190
------------------------------------------------------------------------------
innov | Coef. Std. Err. z P>|z| [95% Conf. Interval]
-------------+----------------------------------------------------------------
er1 | -1.893422 2.764563 -0.68 0.493 -7.311866 3.525022
er2 | .0043424 .0183673 0.24 0.813 -.0316569 .0403417
pro | .9996853 4.239898 0.24 0.814 -7.310363 9.309733
lngov | -.0041834 .1027952 -0.04 0.968 -.2056583 .1972915
lneco | -.1171432 .2046296 -0.57 0.567 -.5182098 .2839233
lnopen | -.1614617 .0601436 -2.68 0.007 -.2793411 -.0435823
lncomp | .2037416 .1538189 1.32 0.185 -.0977378 .5052211
scale | .0085272 .0553505 0.15 0.878 -.0999579 .1170123
_cons | 2.020701 1.33764 1.51 0.131 -.6010259 4.642427
-------------+----------------------------------------------------------------
sigma_u | .09972629
sigma_e | 1.3236944
rho | .00564398 (fraction of variance due to u_i)
------------------------------------------------------------------------------
. hausman FE RE,constant sigmamore
Note: the rank of the differenced variance matrix (8) does not equal the number of coefficients being tested (9); 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.
-------------+----------------------------------------------------------------
er1 | -.1647667 -.074097 -.0906697 .436786
er2 | .0044421 .0039608 .0004813 .0029808
pro | -5.115107 .2616941 -5.376801 5.236186
lngov | -.0687312 .001014 -.0697452 .0946129
lneco | .3032001 .0575211 .245679 .2433602
lnopen | -.0931977 -.0511712 -.0420265 .0318915
lncomp | -.0086796 .0524599 -.0611394 .1580841
scale | -.1341121 -.0705738 -.0635384 .1308041
_cons | -1.79809 -.4100751 -1.388015 1.663905
------------------------------------------------------------------------------
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(8) = (b-B)'[(V_b-V_B)^(-1)](b-B)
= 2.94
Prob>chi2 = 0.9378
(V_b-V_B is not positive definite)