计量经济学消除异方差实验报告.docVIP

计量经济学消除异方差实验报告 姓名：赵国伟 班级：市场营销三班 学号：1014110088 1.建立模型：Y=kX+c Dependent Variable: Y Method: Least Squares Date: 11/13/12 Time: 19:26 Sample: 1 20 Included observations: 20 Variable Coefficient Std. Error t-Statistic Prob.?? C 272.3635 159.6773 1.705713 0.1053 X 0.755125 0.023316 32.38690 0.0000 R-squared 0.983129 ????Mean dependent var 5199.515 Adjusted R-squared 0.982192 ????S.D. dependent var 1625.275 S.E. of regression 216.8900 ????Akaike info criterion 13.69130 Sum squared resid 846743.0 ????Schwarz criterion 13.79087 Log likelihood -134.9130 ????Hannan-Quinn criter. 13.71073 F-statistic 1048.912 ????Durbin-Watson stat 1.189253 Prob(F-statistic) 0.000000 由软件普通最小二乘法统计结果如下： Y=0.755125X+272.3635 R^2=0.983129 F=1048.912 RSS=846743.0 2.下面对该模型进行异方差性检验 已知如果存在异方差性时由X引起的。 首先采用怀特检验结果如下： Heteroskedasticity Test: White F-statistic 23.81765 ????Prob. F(1,18) 0.0001 Obs*R-squared 11.39120 ????Prob. Chi-Square(1) 0.0007 Scaled explained SS 5.013155 ????Prob. Chi-Square(1) 0.0252 Test Equation: Dependent Variable: RESID^2 Method: Least Squares Date: 11/13/12 Time: 19:39 Sample: 1 20 Included observations: 20 Variable Coefficient Std. Error t-Statistic Prob.?? C -5602.797 11961.18 -0.468415 0.6451 X^2 0.001022 0.000209 4.880333 0.0001 R-squared 0.569560 ????Mean dependent var 42337.15 Adjusted R-squared 0.545646 ????S.D. dependent var 45279.67 S.E. of regression 30521.10 ????Akaike info criterion 23.58486 Sum squared resid 1.68E+10 ????Schwarz criterion 23.68444 Log likelihood -233.8486 ????Hannan-Quinn criter. 23.60430 F-statistic 23.81765 ????Durbin-Watson stat 1.031472 Prob(F-statistic) 0.000120 由表知怀特统计量nR^2=11.3912，该值大于5%显著性水平下，自由度为1的χ^2分布的相应临界值3.84，因此拒接同方差的原假设。 3.再采用G-Q检验，将原始数据按X排成升序，去掉中间6个数据，得到两个容量为7的子样本。 对1-7回归得到结果如下 Dependent Variable: Y Method: Least Squares Date: 11/13/12 Time: 20:17 Sample: 1 7 Included observations: 7 Variable Coefficient Std. Error t-Statistic Prob.?? C 1619.577 1936.414 0.836380 0.4411 X 0.483141 0.394049 1.226093 0.