Adjusted R-squared S.E. of regression Sum squared resid Log likelihood Durbin-Watson stat
Coefficient
-1771.254 0.589143 1.433497 0.563954
Std. Error
2530.847 0.080966 0.308466 0.091915
t-Statistic
-0.699866 7.276451 4.647176 6.135601 Prob.
0.4907 0.0000 0.0001 0.0000
125790.9 55317.60 17.67993 17.87025 11245.40 0.000000
0.999289 Mean dependent var 0.999200 S.D. dependent var 1564.382 Akaike info criterion 58734956 Schwarz criterion -243.5191 F-statistic 1.371751 Prob(F-statistic)
(2) 异方差检验 图示法：
从上图可看出，残差e随Y的变动趋势不明显，不规律，所以，该模型可能不存在异方差。是否存在异方差还应通过更进一步的检验。 White检验
White Heteroskedasticity Test: F-statistic
1.042741 Probability 9.595539 Probability
Std. Error
47930201 2913.608 0.046955 0.228951 0.095976 12596.90 0.990310 0.225676 3099.903 0.049458
t-Statistic
-0.600622 0.969097 -0.476773 1.145300 0.146278 0.223609 0.858107 -2.160689 -1.074397 0.471785
Coefficient
-28787936 2823.568 -0.022387 0.262218 0.014039 2816.781 0.849792 -0.487615 -3330.526 0.023334
0.445875 0.384209 Prob.
0.5556 0.3453 0.6393 0.2671 0.8853 0.8256 0.4021 0.0444 0.2968 0.6427
Obs*R-squared
Test Equation:
Dependent Variable: RESID^2 Method: Least Squares Date: 5/21/13 Time:11:13 Sample: 1980 xx Included observations: 28
Variable C X1 X1^2 X1*X3 X1*X4 X3 X3^2 X3*X4 X4 X4^2
R-squared
Adjusted R-squared S.E. of regression Sum squared resid Log likelihood Durbin-Watson stat
0.342698 Mean dependent var 0.014047 S.D. dependent var 2715618. Akaike info criterion 1.33E+14 Schwarz criterion -448.3515 F-statistic 3.175863 Prob(F-statistic)
2097677. 2734894. 32.73939 33.21518 1.042741 0.445875 2
nR2=9.595539,由White检验知，在α=0.05下，查χ2分布表，得临界值χ因为nR2=9.595539<χ不存在异方差。
ARCH检验：
ARCH Test: F-statistic
0.731099 Probability 0.767152 Probability
Std. Error
679705.5 0.196543
t-Statistic
3.542855 -0.855043 Prob. 20.05 0.05
(10)=18.3070。
(10)=18.3070。所以拒绝备择假设，不拒绝原假设，表明模型
Obs*R-squared
财政学毕业论文选题
