1、485.88539.521.075496.56576.721.081520.84604.311.086599.64728.171.106770.64875.521.25949.081069.611.3361071.041187.491.4261278.871329.71.6671291.091477.771.9121440.471638.921.971585.711844.982.1711907.712238.382.4182322.192769.262.8443301.373982.133.5264064.14929.534.0664679.615967.714.4325204.296608
2、.564.5695471.017110.544.5465851.537649.834.4966121.078140.554.478 4、GENR Y=CONSUM /PRICE 回车 5、GENR X= INCOME /PRICE 回车6、SCAT X Y 回车2)相关图分析 Scat x y,得到关于X和Y的散点图如下 从上图可知,X和Y存在线性关系。7、LS Y C X 回车 Eviews 估计结果如下图:Dependent Variable: YMethod: Least SquaresDate: 05/25/13 Time: 10:59Sample: 1978 2000Included
3、 observations: 23VariableCoefficientStd. Errort-StatisticProb. C111.449617.059026.533181X0.7118290.01690242.11456R-squared0.988298 Mean dependent var769.4132Adjusted R-squared0.987741 S.D. dependent var296.7211S.E. of regression32.85273 Akaike info criterion9.904888Sum squared resid22665.35 Schwarz
4、criterion10.00363Log likelihood-111.906 F-statistic1773.636Durbin-Watson stat0.598569 Prob(F-statistic)估计线性回归模型并计算残差用普通最小二乘法求估计的回归方程结果如下(6.5) (42.1) R2 =0.9883 s.e=32.8 DW=0.60 T=23回归方程拟合得效果比较好,但是DW值比较低。3)自相关检验 1)图示法 LINR RESID; SCAT RESID(-1) RESID; 2)观察结果窗口,由DW统计量,查表,与DL,DU比较得出结论;1)图示法: 由scat resi
5、d(-1) resid 得到下图:由上图可知,方程存在序列相关性。2)已知DW=0.60,若给定a=0.05,查表得DW临界值 dL=1.26,dU=1.44 。因为DW=0.061.26,认为误差项存在严重自相关。3)LM检验 在方程窗口中点击Viewresidual test series correlation LM test;Breusch-Godfrey Serial Correlation LM Test:F-statistic7.584071Probability0.003791Obs*R-squared10.210310.006065Test Equation: RESID 1
6、1:00Presample missing value lagged residuals set to zero.Prob.3.90350913.386240.2916060.7737-0.0056610.013323-0.4249120.6757RESID(-1)0.5966480.2317782.5742270.0186RESID(-2)0.1405640.2365540.5942150.55940.443926Mean dependent var-2.32E-140.356125S.D. dependent var31.7664125.48994Akaike info criterion
7、9.47121512345.00Schwarz criterion9.668693-104.91905.0560471.765655Prob(F-statistic)0.009645LMBG自相关检验辅助回归方程式估计结果是: 3.9) 0.2 -0.4 R2 =0.43 DW=2.00 LM= =230.43=9.89因为 ,LM=9.893.84,所以LM检验结果也说明误差项存在一阶正相关。(4)自相关的修正 GENR GDY=Y-0.7*Y(-1); GENR GDX=X-0.7*X(-1); LS GDY C GDX; ; GDY04Sample (adjusted): 1979 20
8、00 22 after adjustments46.5422111.876453.9188640.0009GDX0.6746730.03282720.552380.00000.954792269.78440.952532103.391422.526119.15373510148.519.252921-98.69108422.40052.3104240.000000 令GDY=Y-0.70*Y(-1) GDX=X-0.70X(-1) 以GDY、GDX、(1979-2000)为样本再次回归,得 GDY=45.2489+0.6782GDX (3.7) (20.0) R2 =0.95 s.e=23.2
9、 DW=2.31, T=22(19792000) 回归方程拟合优度依然较好,且DW=2.13.查表得,。因为 DW=2,132.57,依据判断规则,误差项已经消除自相关。 (5)再次检验自相关是否存在,用1),2),3)之一检验0.6608060.5285211.5048160.471230070.07870712.089110.0065110.9949-0.0002740.033404-0.0081990.9935-0.1361810.232983-0.5845130.56610.2047850.2330330.8787800.39110.0684013.42E-14-0.08686621.9832322.918149.2647019454.3459.463072-97.911710.4405371.9406150.726816由上图知 D.W约为2,误差项已消除自相关。 五、 实验数据记录、处理及结果分析 6、讨论、心得1、通过本次试验,我进一步熟悉了掌握了Eviews软件的操作,并且掌握了自相关的检验以及修正方法。我认为在输出图像,图表的同时,分析解释非常重要。保持一颗细心,耐心的态度,是做好事情的基础。也锻炼了我的严谨的能力,我相信我一定会把计量经济学学好,相信自己行,一定我能行! 附件三:实验报告附页山东轻工业学院实验报告(附页)
copyright@ 2008-2022 冰豆网网站版权所有
经营许可证编号:鄂ICP备2022015515号-1