计量经济学模型检验作业答案(共4页).doc

《计量经济学模型检验作业答案(共4页).doc》由会员分享，可在线阅读，更多相关《计量经济学模型检验作业答案(共4页).doc（4页珍藏版）》请在淘文阁 - 分享文档赚钱的网站上搜索。

1、精选优质文档-倾情为你奉上y=c(1)+c(2)*x8、下表列出了某年中国部分省市城镇居民每个家庭平均全年可支配收入X与消费性支出Y的统计数据。obsXY200110349.6910349.6920028140.58140.520035129.055129.0520045357.795357.7920054810481020064912.884912.8820075661.165661.1620084724.114724.1120094766.264766.2620105524.545524.5420116218.736218.7320129761.579761.57201311718.011

2、1718.0120146800.236800.2320159279.169279.1620166489.976489.9720175124.245124.2420184916.254916.2520195169.965169.9620205644.865644.86散点图如图所示：（1）线性分析如表所示：Dependent Variable: YMethod: Least SquaresDate: 11/22/11 Time: 21:56Sample: 2001 2020Included observations: 20Y=C(1)+C(2)*XCoefficientStd. Errort-S

3、tatisticProb.C(1)272.3635159.67731.0.1053C(2)0.0.32.386900.0000R-squared0.Mean dependent var5199.515Adjusted R-squared0.S.D. dependent var1625.275S.E. of regression216.8900Akaike info criterion13.69130Sum squared resid.0Schwarz criterion13.79087Log likelihood-134.9130Durbin-Watson stat1.（2）异方差性检验首先采

4、用GQ检验。在对20个样本按x从大到小排序，去掉中间4个个体，对前后两个样本进行OLS估计，样本容量均为8.前一个样本的估计结果如表所示：Dependent Variable: YMethod: Least SquaresDate: 11/22/11 Time: 22:17Sample: 2001 2008Included observations: 8Y=C(1)+C(2)*XCoefficientStd. Errort-StatisticProb.C(1)212.2118530.88920.0.7032C(2)0.0.12.625050.0000R-squared0.Mean depend

5、ent var6760.477Adjusted R-squared0.S.D. dependent var1556.814S.E. of regression320.2790Akaike info criterion14.58858Sum squared resid.0Schwarz criterion14.60844Log likelihood-56.35432Durbin-Watson stat1.后一个样本的估计量如表所示：Dependent Variable: YMethod: Least SquaresDate: 11/22/11 Time: 22:28Sample: 2001 20

6、08Included observations: 8Y=C(1)+C(2)*XCoefficientStd. Errort-StatisticProb.C(1)1277.1611540.6040.0.4388C(2)0.0.1.0.1255R-squared0.Mean dependent var4016.814Adjusted R-squared0.S.D. dependent var166.1712S.E. of regression145.2172Akaike info criterion13.00666Sum squared resid.3Schwarz criterion13.02652Log likelihood-50.02663Durbin-Watson stat3.于是得到如下F统计量：F=RSS1/RSS2=.0/.3=4.86在5%的显著性水平下，自由度（6,6）的F分布的临界值为F0.05(6,6)=4.28。于是拒绝无异方差性检验，表明原模型存在异方差性。 其次，采用White检验。在对原模型进行OLS估计后，得到如图琐事结果：F-statistic14.63595Probability0.Obs*R-squared12.65213Probability0.从伴随概率值容易看出，在5%的显著性水平下，原模型存在异方差性。专心-专注-专业