Dependent Variable: Y Method: Least Squares Date: 12/16/14 Time: 19:41 Sample: 1978 2012 Included observations: 35
Variable C X
R-squared
Coefficient
-3348.918 0.214508
Std. Error
848.4344 0.004706
t-Statistic
-3.947174 45.57942
Prob.
0.0004 0.0000
21015.69 30709.83 19.42969 19.51856 2077.483 0.000000
0.984364 Mean dependent var 0.983890 S.D. dependent var 3897.858 Akaike info criterion 5.01E+08 Schwarz criterion -338.0195 F-statistic 0.122825 Prob(F-statistic)
Adjusted R-squared S.E. of regression Sum squared resid Log likelihood Durbin-Watson stat
图1 回归结果
ˆ=-3348.918+0.214508X Y i i
t =(-3.9472) (45. 579 4
3DW =0. 122R 2=0.9844 2=0. 983
8 F =2077. 4 8 8
图2 剩余项、实际值、拟合值图形
图3 散点图
图4 趋势图
White Heteroskedasticity Test: F-statistic
10.08252 Probability 13.52968 Probability
Std. Error
3129007.
t-Statistic
4.193653
Coefficient
13121967
0.000402 0.001154
Prob.
0.0002
Obs*R-squared
Test Equation:
Dependent Variable: RESID^2 Method: Least Squares Date: 12/16/14 Time: 19:49 Sample: 1978 2012 Included observations: 35
Variable C
X X^2
R-squared
-67.36311 0.000272
47.03735 0.000102
-1.432119 2.661806
0.1618 0.0121
14325106 14715600 35.50048 35.63379 10.08252 0.000402
0.386562 Mean dependent var 0.348223 S.D. dependent var 11880307 Akaike info criterion 4.52E+15 Schwarz criterion -618.2584 F-statistic 0.689279 Prob(F-statistic)
Adjusted R-squared S.E. of regression Sum squared resid Log likelihood Durbin-Watson stat
图5 White检验结果
由图5可知，当取α=0.05时，nR 2=13.52968
2χ0.05(2)=5.9915，由于
2
nR 2=13.52968>χ0.05(2)=5.9915, 同时p 值为0.001154，小于0.05的显著性水平，表明
模型中存在异方差性。
表1 国内生产总值及财政收入 单位：亿元
资料来源：《中国统计年鉴2013》 散点图：scat x y 趋势图：plot y x
生成数据：data y x 回归：ls y c x scat x y plot y x
Dependent Variable: Y Method: Least Squares Date: 12/16/14 Time: 19:41 Sample: 1978 2012 Included observations: 35
Variable C X
R-squared
Coefficient
-3348.918 0.214508
Std. Error
848.4344 0.004706
t-Statistic
-3.947174 45.57942
Prob.
0.0004 0.0000
21015.69 30709.83 19.42969 19.51856 2077.483 0.000000
0.984364 Mean dependent var 0.983890 S.D. dependent var 3897.858 Akaike info criterion 5.01E+08 Schwarz criterion -338.0195 F-statistic 0.122825 Prob(F-statistic)
Adjusted R-squared S.E. of regression Sum squared resid Log likelihood Durbin-Watson stat
图1 回归结果
ˆ=-3348.918+0.214508X Y i i
t =(-3.9472) (45. 579 4
3DW =0. 122R 2=0.9844 2=0. 983
8 F =2077. 4 8 8
图2 剩余项、实际值、拟合值图形
图3 散点图
图4 趋势图
White Heteroskedasticity Test: F-statistic
10.08252 Probability 13.52968 Probability
Std. Error
3129007.
t-Statistic
4.193653
Coefficient
13121967
0.000402 0.001154
Prob.
0.0002
Obs*R-squared
Test Equation:
Dependent Variable: RESID^2 Method: Least Squares Date: 12/16/14 Time: 19:49 Sample: 1978 2012 Included observations: 35
Variable C
X X^2
R-squared
-67.36311 0.000272
47.03735 0.000102
-1.432119 2.661806
0.1618 0.0121
14325106 14715600 35.50048 35.63379 10.08252 0.000402
0.386562 Mean dependent var 0.348223 S.D. dependent var 11880307 Akaike info criterion 4.52E+15 Schwarz criterion -618.2584 F-statistic 0.689279 Prob(F-statistic)
Adjusted R-squared S.E. of regression Sum squared resid Log likelihood Durbin-Watson stat
图5 White检验结果
由图5可知，当取α=0.05时，nR 2=13.52968
2χ0.05(2)=5.9915，由于
2
nR 2=13.52968>χ0.05(2)=5.9915, 同时p 值为0.001154，小于0.05的显著性水平，表明
模型中存在异方差性。
表1 国内生产总值及财政收入 单位：亿元
资料来源：《中国统计年鉴2013》 散点图：scat x y 趋势图：plot y x
生成数据：data y x 回归：ls y c x scat x y plot y x