《实验三SPSS多元时间序列分析方法.doc》由会员分享，可在线阅读，更多相关《实验三SPSS多元时间序列分析方法.doc（23页珍藏版）》请在三一文库上搜索。
1、实验三 多元时间序列分析方法1. 实验目的了解协整理论及协整检验方法；掌握协整的两种检验方法：E-G两步法与Johansen方法；熟悉向量自回归模型VAR的应用；掌握误差修正模型ECM的含义及检验方法；掌握Granger因果关系检验方法。2. 实验仪器装有EViews7.0软件的微机一台。3. 实验内容【例6-2】时间与M2之间的关系首先用单位根检验是否为平稳序列。原假设为H0：非平稳序列 H1：平稳序列。用Eviews软件解决该问题，得到如下结果：Null Hypothesis: M2 has a unit rootExogenous: NoneLag Length: 3 (Automati
2、c - based on SIC, maxlag=13)t-StatisticProb.*Augmented Dickey-Fuller test statistic5.6811691.0000Test critical values:1% level-2.5790525% level-1.94276810% level-1.615423*MacKinnon (1996) one-sided p-values.Augmented Dickey-Fuller Test EquationDependent Variable: D(M2)2 / 23Method: Least SquaresDate
3、: 04/16/13 Time: 10:36Sample (adjusted): 1991M05 2005M01Included observations: 165 after adjustmentsVariableCoefficientStd. Errort-StatisticProb.M2(-1)0.0135140.0023795.6811690.0000D(M2(-1)-0.4902800.074458-6.5846110.0000D(M2(-2)0.0706180.0837900.8427970.4006D(M2(-3)0.3870860.0737885.2459350.0000R-s
4、quared0.480147Mean dependent var1440.037Adjusted R-squared0.470461S.D. dependent var1509.489S.E. of regression1098.447Akaike info criterion16.86513Sum squared resid1.94E+08Schwarz criterion16.94042Log likelihood-1387.373Hannan-Quinn criter.16.89569Durbin-Watson stat1.965242从上图我们可以看出t-statistic的值是5.6
5、81169，大于临界值，pa，故不能拒绝被检验的指数序列是非平稳的原假设。因此一阶差分序列进行ADF检验，结果如下图显示。Null Hypothesis: D(M2) has a unit rootExogenous: NoneLag Length: 8 (Automatic - based on SIC, maxlag=13)t-StatisticProb.*Augmented Dickey-Fuller test statistic0.9881830.9143Test critical values:1% level-2.5795875% level-1.94284310% level-1
6、.615376*MacKinnon (1996) one-sided p-values.Augmented Dickey-Fuller Test EquationDependent Variable: D(M2,2)Method: Least SquaresDate: 04/16/13 Time: 10:37Sample (adjusted): 1991M11 2005M01Included observations: 159 after adjustmentsVariableCoefficientStd. Errort-StatisticProb.D(M2(-1)0.0536160.0542
7、570.9881830.3247D(M2(-1),2)-1.5260690.096352-15.838520.0000D(M2(-2),2)-1.5196490.149134-10.189810.0000D(M2(-3),2)-1.2256230.184003-6.6608690.0000D(M2(-4),2)-1.2374450.196285-6.3043190.0000D(M2(-5),2)-0.9720240.197161-4.9300930.0000D(M2(-6),2)-0.8100980.185290-4.3720600.0000D(M2(-7),2)-0.6050690.1449
8、97-4.1729830.0001D(M2(-8),2)-0.3337810.080550-4.1437810.0001R-squared0.801713Mean dependent var16.07001Adjusted R-squared0.791137S.D. dependent var2352.919S.E. of regression1075.320Akaike info criterion16.85356Sum squared resid1.73E+08Schwarz criterion17.02727Log likelihood-1330.858Hannan-Quinn crit
9、er.16.92410Durbin-Watson stat1.970407从上图我们可以看出t-statistic的值是0.988183，大于临界值，pa，故不能拒绝被检验的指数序列是非平稳的原假设。因此二阶差分序列进行ADF检验，结果如下图显示Null Hypothesis: D(M2,2) has a unit rootExogenous: NoneLag Length: 7 (Automatic - based on SIC, maxlag=13)t-StatisticProb.*Augmented Dickey-Fuller test statistic-9.2231320.0000T
10、est critical values:1% level-2.5795875% level-1.94284310% level-1.615376*MacKinnon (1996) one-sided p-values.Augmented Dickey-Fuller Test EquationDependent Variable: D(M2,3)Method: Least SquaresDate: 04/16/13 Time: 10:38Sample (adjusted): 1991M11 2005M01Included observations: 159 after adjustmentsVa
11、riableCoefficientStd. Errort-StatisticProb.D(M2(-1),2)-8.9007550.965047-9.2231320.0000D(M2(-1),3)6.4311290.9246726.9550380.0000D(M2(-2),3)4.9702860.8335415.9628610.0000D(M2(-3),3)3.8024320.7007735.4260550.0000D(M2(-4),3)2.6170580.5445964.8055010.0000D(M2(-5),3)1.6882010.3805594.4361090.0000D(M2(-6),
12、3)0.9109680.2149904.2372570.0000D(M2(-7),3)0.3259340.0801514.0664870.0001R-squared0.941321Mean dependent var0.112057Adjusted R-squared0.938601S.D. dependent var4339.324S.E. of regression1075.236Akaike info criterion16.84747Sum squared resid1.75E+08Schwarz criterion17.00188Log likelihood-1331.374Hann
13、an-Quinn criter.16.91018Durbin-Watson stat1.963915从上图我们可以看出t-statistic的值是-9.223132，小于临界值，pa，故不能拒绝被检验的指数序列是非平稳的原假设。因此一阶差分序列进行ADF检验，结果如下图显示。Null Hypothesis: D(X) has a unit rootExogenous: NoneLag Length: 0 (Automatic - based on SIC, maxlag=20)t-StatisticProb.*Augmented Dickey-Fuller test statistic-29.
14、946780.0000Test critical values:1% level-2.5675785% level-1.94118110% level-1.616459*MacKinnon (1996) one-sided p-values.Augmented Dickey-Fuller Test EquationDependent Variable: D(X,2)Method: Least SquaresDate: 04/16/13 Time: 10:54Sample (adjusted): 1/06/2003 9/14/2006Included observations: 885 afte
15、r adjustmentsVariableCoefficientStd. Errort-StatisticProb.D(X(-1)-1.0071650.033632-29.946780.0000R-squared0.503597Mean dependent var-1.13E-05Adjusted R-squared0.503597S.D. dependent var0.063302S.E. of regression0.044600Akaike info criterion-3.381041Sum squared resid1.758410Schwarz criterion-3.375633
16、Log likelihood1497.111Hannan-Quinn criter.-3.378974Durbin-Watson stat1.999816从上图我们可以看出t-statistic的值是-29.94678，远小于临界值，pa，故拒绝被检验的指数序列是平稳的原假设。Dependent Variable: XMethod: Least SquaresDate: 04/16/13 Time: 11:00Sample (adjusted): 1/03/2003 9/14/2006Included observations: 886 after adjustmentsConvergence
17、 achieved after 4 iterationsVariableCoefficientStd. Errort-StatisticProb.C2.4667710.20897311.804270.0000AR(1)0.9926970.003809260.60880.0000R-squared0.987151Mean dependent var2.386998Adjusted R-squared0.987137S.D. dependent var0.392414S.E. of regression0.044506Akaike info criterion-3.384131Sum square
18、d resid1.751013Schwarz criterion-3.373325Log likelihood1501.170Hannan-Quinn criter.-3.380000F-statistic67916.95Durbin-Watson stat2.008279Prob(F-statistic)0.000000Inverted AR Roots.99四、实验心得通过这次的实验我感觉自己的水平还是很有限的，感觉实际操作中有很多的不足，课后自己也重新试过，感觉有很多还是没记住，但是在这次的实验中自己也有懂得了更多 友情提示：方案范本是经验性极强的领域，本范文无法思考和涵盖全面，供参考!最好找专业人士起草或审核后使用。