the final mle estimates are :
coefficient standard-error t-ratio
beta 0 -0.62732292E+01 0.26163916E+00 -0.23976644E+02
beta 1 0.34852557E+00 0.94923735E-01 0.36716378E+01
beta 2 -0.55850492E+00 0.99783346E-01 -0.55971757E+01
beta 3 0.32531683E+00 0.74566451E-01 0.43627775E+01
beta 4 0.30500471E+00 0.72378104E-01 0.42140466E+01
beta 5 0.15925062E+00 0.74086634E-01 0.21495189E+01
beta 6 0.43346044E+00 0.76580768E-01 0.56601735E+01
delta 1 0.64435526E+01 0.12380347E+01 0.52046622E+01
delta 2 0.54740793E+01 0.27983530E+01 0.19561790E+01
delta 3 -0.77174992E-03 0.18752374E-03 -0.41154786E+01
delta 4 0.12511003E+00 0.88146286E-01 0.14193455E+01
delta 5 -0.19814736E-01 0.13211451E-01 -0.14998153E+01
sigma-squared 0.33157699E+00 0.45914452E-01 0.72216258E+01
gamma 0.73839699E+00 0.62067712E-01 0.11896636E+02
log likelihood function = -0.15847098E+03
LR test of the one-sided error = 0.71225560E+02
with number of restrictions = 6
[note that this statistic has a mixed chi-square distribution]
number of iterations = 23
(maximum number of iterations set at : 100)
number of cross-sections = 30
number of time periods = 11
total number of observations = 330
thus there are: 0 obsns not in the panel
一般文献中都会给回归系数标上在10%,5%,1%的置信水平下的显著性,就是用* ** ***
请问这个是根据t检验值判断的?
如果是,t检验值怎么计算?
不会就是t-ratio吧?那我的结果岂不是都不显著!哭瞎。。。


雷达卡





京公网安备 11010802022788号







