得到结果如下：
Dependent Variable: CONSUME? Method: Pooled Least Squares Date: 07/21/14 Time: 15:44 Sample: 1996 2002 Included observations: 7 Cross-sections included: 15
Total pool (balanced) observations: 105
Variable C INCOME? Fixed Effects (Cross)
AH--C BJ--C FJ--C HB--C HLJ--C JL--C JS--C JX--C LN--C NMG--C SD--C SH--C
Coefficient 806.6751 0.653338 -94.50854 698.0132 -18.86465 -200.3997 -246.3712 -54.16421 -31.26919 -392.9844 47.39508 -284.2660 -150.8912 465.4906
11
Std. Error 221.2143 0.034541
t-Statistic 3.646578 18.91504
Prob. 0.0005 0.0000
SX--C TJ--C ZJ--C
Fixed Effects (Period)
1996--C 1997--C 1998--C 1999--C 2000--C 2001--C 2002--C
-152.6560 103.9569 311.5193 -59.12373 17.95469 -31.45564 -57.24042 36.24382 -29.26415 122.8854
4981.017 1700.985 13.12288 13.67895 13.34821 1.455623
Effects Specification
Cross-section fixed (dummy variables) Period fixed (dummy variables) R-squared Adjusted R-squared S.E. of regression Sum squared resid Log likelihood F-statistic Prob(F-statistic)
0.993278 Mean dependent var 0.991577 S.D. dependent var 156.1067 Akaike info criterion 2022652. Schwarz criterion -666.9514 Hannan-Quinn criter. 584.0406 Durbin-Watson stat 0.000000
接下来用F统计量检验是应该建立混合回归模型，还是个体固定效应回归模型。
H0：?1=?2=?????N?1和?1=?2=???=?T?1?0： 对模型进行检验：
(RRSS?URSS)（4965275-2022652）（T?N?2）22-2=5.83?F（20,83）F?==17. 0.05URSS2022652(NT?T?N?K?1)83所以推翻原假设，可以建立个体时点固定效应回归模型
12