计量经济学课程论文——影响我国财政收入因素的实证分析(3)

R-squared Adjusted R-squared S.E. of regression Sum squared resid Log likelihood Coefficient 269.4737 0.314806 0.012259 0.190484

Std. Error 327.9155 0.135305 0.008358 0.059992

t-Statistic 0.821778 2.326643 1.466691 3.175158

Prob. 0.4184 0.0277 0.1540 0.0037 1503.341 2477.187 12.42169 12.60672 12.48200

0.693436 -0.022309 0.124056

0.102698 0.007573 0.030513

6.752167 -2.945694 4.065674

0.0000 0.0066 0.0004 3109.770 766.5424 13.70148 13.88651 13.76179 1.141270

16520.73 22001.07 25761939

Weighted Statistics

0.990514 Mean dependent var 0.989460 S.D. dependent var 215.2590 Akaike info criterion 1251084. Schwarz criterion -208.3729 Hannan-Quinn criter. 939.7561 Durbin-Watson stat 0.000000

Unweighted Statistics

0.998226 Mean dependent var 0.998029 S.D. dependent var 976.8039 Sum squared resid 0.595311

Weighted Statistics

0.935786 Mean dependent var 0.928651 S.D. dependent var 113.5166 Akaike info criterion 347922.4 Schwarz criterion -188.5362 Hannan-Quinn criter.

影响我国财政收入因素的实证分析计量经济学论文,学计量的人,伤不起啊!!!

F-statistic Prob(F-statistic)

R-squared Adjusted R-squared S.E. of regression Durbin-Watson stat

131.1561 Durbin-Watson stat 0.000000

Unweighted Statistics

1.696653 16520.73 22001.07 3.80E+08

0.973838 Mean dependent var 0.970931 S.D. dependent var 3751.096 Sum squared resid 0.106940

Dependent Variable: Y Method: Least Squares Date: 12/14/11 Time: 14:29 Sample: 1980 2010 Included observations: 31 Weighting series: W3

Variable C X2 X3 X4

R-squared Adjusted R-squared S.E. of regression Sum squared resid Log likelihood F-statistic Prob(F-statistic)

R-squared Adjusted R-squared S.E. of regression Durbin-Watson stat

Coefficient 2179.146 0.988129 -0.038288 0.045264

Std. Error 265.3987 0.053669 0.005657 0.015336

t-Statistic 8.210839 18.41153 -6.768766 2.951390

Prob. 0.0000 0.0000 0.0000 0.0065 6966.490 4843.265 14.04105 14.22609 14.10137 1.238294

16520.73 22001.07 3186080.

Weighted Statistics

0.999006 Mean dependent var 0.998895 S.D. dependent var 255.0939 Akaike info criterion 1756968. Schwarz criterion -213.6363 Hannan-Quinn criter. 9043.425 Durbin-Watson stat 0.000000

Unweighted Statistics

0.999781 Mean dependent var 0.999756 S.D. dependent var 343.5156 Sum squared resid 1.099033

经比较可知，用权数w3的效果最好，修正后的模型为：

影响我国财政收入因素的实证分析计量经济学论文,学计量的人,伤不起啊!!!

t=2179.146+0.988129X2-0.038288X3+0.045264X4 Y

t=（8.210839）（18.41153）（-6.768766）（2.951390） R2=0.999006 R2=0.998895 F=9043.425 DW=1.238294

7、自相关的检验及其修正

7.1自相关的检验

1%的显著性水平，查DW统计表可知，dL=0.960 dU1.510。此模型DW值为1.238294。为无法判断的区域。此时只能改用图示法来检验。

通过EViews软件得出该模型的残差图如下

上图表明模型显然存在自相关。在这里我们使用广义差分法进行弥补。 7.2自相关修正

Dependent Variable: E Method: Least Squares Date: 12/14/11 Time: 14:47 Sample (adjusted): 1981 2010

Included observations: 30 after adjustments

Variable E(-1)

R-squared Adjusted R-squared S.E. of regression

Coefficient 0.420667

Std. Error 0.221680

t-Statistic 1.897630

Prob. 0.0677 0.114935 331.4580 14.36060

0.110457 Mean dependent var 0.110457 S.D. dependent var 312.6166 Akaike info criterion

影响我国财政收入因素的实证分析计量经济学论文,学计量的人,伤不起啊!!!

Sum squared resid Log likelihood Durbin-Watson stat

2834145. Schwarz criterion -214.4090 Hannan-Quinn criter. 1.188304

14.40730 14.37554

e t=0.420667et-1

由此可知ρ=0.420667，对原模型进行广义差分，得到广义差分方程为：

Yt-0.420667Yt-1=β0(1-0.420667)+β2(X2t-0.420667X2t-1)+β

3(X3t-0.420667X3t-1)+β4(X4t-0.420667X4t-1) 对广义差分方程进行回归

Dependent Variable: Y-0.420667*Y(-1) Method: Least Squares Date: 12/14/11 Time: 15:10 Sample (adjusted): 1981 2010

Included observations: 30 after adjustments

Variable C

X2-0.420667*X2(-1) X3-0.420667*X3(-1) X4-0.420667*X4(-1) R-squared Adjusted R-squared S.E. of regression Sum squared resid Log likelihood F-statistic Prob(F-statistic)

Coefficient 984.5681 1.059967 -0.032215 0.023950

Std. Error 399.3872 0.029968 0.011536 0.007758

t-Statistic 2.465197 35.36950 -2.792647 3.087177

Prob. 0.0206 0.0000 0.0097 0.0048 11016.64 14422.34 14.20564 14.39247 14.26541 1.755276

0.999671 Mean dependent var 0.999633 S.D. dependent var 276.4690 Akaike info criterion 1987312. Schwarz criterion -209.0846 Hannan-Quinn criter. 26297.38 Durbin-Watson stat 0.000000

t*=984.5681+1.059967X2*-0.032215X3*+0.023950X4* Y

t=（2.465197）（35.36950）（-2.792647）（3.087177） R2=0.999671 R2=0.999633 F=26297.38 DW=1.755276

样本容量减少一个，为30个，1%的显著性水平，查DW统计表克制，dL=0.941 dU=1.511.模型中DW=1.755276> dU 这表明在1%的显著性水平下广义差分模型中已经无自相关，同时R2、t、F统计量也都达到理想水平。 由差分方程有

984.5681

β=1 0.420667

影响我国财政收入因素的实证分析计量经济学论文,学计量的人,伤不起啊!!!

t=1699.4856+1.059967X2-0.032215X3+0.023950X4 Y

t=（2.465197）（35.36950）（-2.792647）（3.087177） R2=0.999671 R2=0.999633 F=26297.38 DW=1.755276

8、结论与对策

8.1结论：

1，从模型可以看出，在我国，税收收入与财政收入存在着高度的正相关，税收收入的增长对财政收入的增长有重大的促进作用；

2，全国从业 …… 此处隐藏：2962字，全部文档内容请下载后查看。喜欢就下载吧 ……