Dependent Variable: Y Method: Least Squares Date: 12/24/10 Time: 12:38 Sample: 1989 2008 Included observations: 20
C X1
R-squared
Adjusted R-squared S.E. of regression Sum squared resid Log likelihood Durbin-Watson stat
500.3204 0.494455
87.09007 0.028289
5.744861 17.47869
0.0000 0.0000 908.3138 13.72092 13.82050 305.5045 0.000000
0.944359 Mean dependent var 1756.100 0.941268 S.D. dependent var 220.1266 Akaike info criterion 872202.8 Schwarz criterion -135.2092 F-statistic 0.367926 Prob(F-statistic)
将Y与X2做回归得到结果如表4：
Dependent Variable: Y Method: Least Squares Date: 12/24/10 Time: 12:40 Sample: 1989 2008 Variable C X2
R-squared
Adjusted R-squared S.E. of regression Sum squared resid Log likelihood Durbin-Watson stat
Coefficient 6964.538 -49.80339
Std. Error 2894.609 27.61744
t-Statistic 2.406038 -1.803331
Prob. 0.0271 0.0881
0.153021 Mean dependent var 1756.100 0.105967 S.D. dependent var 858.8410 Akaike info criterion 13276940 Schwarz criterion -162.4368 F-statistic 0.187833 Prob(F-statistic)
908.3138 16.44368 16.54326 3.252004 0.088102
计算各解释变量的相关系数，选择X1、X2的数据，得到相关系数矩阵如表5：
变量 X1 X2
X1 1.000000 -0.298504
X2 -0.298504 1.000000
由表3和表4可知，Y与X1的组合为最优方程，虽然X2与Y的拟合度不