时间序列分析VAR模型实验Word文件下载.docx

1、LeastSquaresDate:11/15/14Time:20:20Sample（adjusted）:2005M112014M10Includedobservations:108afteradjustmentsVariableCoefficientStd.ErrorProb.DLE（-1）-0.3530050.1164390.0031D（DLE（-1）-0.5027300.115417-4.3557680.0000D（DLE（-2）-0.3115310.093265-3.3402580.0012C-0.0008880.000470-1.8875920.0619R-squared0.45024

2、0Meandependentvar1.15E-05AdjustedR-squared0.434382S.D.dependentvar0.005058S.E.ofregression0.003804Akaikeinfocriterion-8.269046Sumsquaredresid0.001505Schwarzcriterion-8.169708Loglikelihood450.5285Hannan-Quinncriter.-8.228768F-statistic28.39119Durbin-Watsonstat2.061613Prob（F-statistic）0.000000单位根统计量AD

3、F=-3.031673小于临界值，且P为0.0351，因此该序列不是单位根过程，即该序列是平稳序列。4.1.2国房景气指数P序列首先作出P序列的时序图：图4.4P的曲线图由于每年一月份的数据缺失，故取相邻两项进行平均补全数据，得到新序列的时序图如下：图4.5P的曲线图（补全）由上图可知，该序列P可能存在一定的趋势性和季节性，先进行单位根检验，确定改序列是否平稳。由于序列表4.2单位根输出结果PhasaunitrootConstant,LinearTrend3（Automatic-basedonSIC,maxlag=12）-3.9724570.0124-4.045236-3.451959-3.1

4、51440由单位根检验结果可知，T值小于临界值，且P=0.0124，在5%的置信水平下，该序列不存在单位根过程。由于汇率E序列为一阶单整序列，并进行了一阶差分处理，因此样本数量减少，在下面的操作中，所有的样本序列调整为2005-08至2014-10。4.2模型参数识别先进行VAR模型的拟合，初步选定滞后阶数为3：表4.3拟合输出结果VectorAutoregressionEstimates11/22/14Time:22:Standarderrorsin（）&t-statisticsinDLEP0.063183-19.12274（0.09626）（14.1374）0.65638-1.35263D

5、LE（-2）0.11679815.42129（0.09604）（14.1052）1.216151.09330DLE（-3）0.24526016.39171（0.09617）（14.1243）2.550301.16053P（-1）-9.04E-051.490708（0.00066）（0.09765）-0.1359315.2656P（-2）-0.000583-0.355442（0.00118）（0.17380）-0.49226-2.04508P（-3）0.000346-0.160740（0.00067）（0.09872）0.51479-1.628210.0313282.571540（0.01274

6、）（1.87084）2.459431.374540.2950330.979509Adj.R-squared0.2531540.978292Sumsq.resids0.00139029.99247S.E.equation0.0037100.5449367.044848804.6767454.8094-84.06138AkaikeAIC-8.2927661.686322SchwarzSC-8.1189241.860164Meandependent-0.002527100.2406S.D.dependent0.0042933.698585Determinantresidcovariance（dofa

7、dj.）4.08E-06Determinantresidcovariance3.57E-06370.8871Akaikeinformationcriterion-6.609021-6.261337再进行滞后阶数的确定：表4.4最优滞后阶数的判断VARLagOrderSelectionCriteriaEndogenousvariables:DLEPExogenousvariables:22Sample:2005M072014M1099LagLogLLRFPEAICSCHQ134.7784NA0.000234-2.682392-2.629965-2.6611801302.5627325.39998

8、.57e-06-5.991165-5.833886-5.9275302329.023050.247835.45e-06-6.444909-6.182775*-6.338849*3334.37339.943949*5.30e-06*-6.472187*-6.105200-6.3237044337.45315.5997425.40e-06-6.453598-5.981758-6.2626915339.75894.0991765.60e-06-6.419372-5.842679-6.1860416345.04989.1923245.46e-06-6.445451-5.763905-6.1696967

9、345.54840.8460765.87e-06-6.374716-5.588316-6.0565378346.73691.9687606.23e-06-6.317917-5.426663-5.9573149352.58019.4436396.01e-06-6.355154-5.359047-5.95212810353.77141.8770826.39e-06-6.298411-5.197451-5.85296111354.36490.9112796.87e-06-6.229594-5.023780-5.74172012356.46173.1346447.18e-06-6.191146-4.8

10、80479-5.660848*indicateslagorderselectedbythecriterionLR:sequentialmodifiedLRteststatistic（eachtestat5%level）FPE:FinalpredictionerrorAIC:SC:SchwarzinformationcriterionHQ:Hannan-Quinninformationcriterion由上边可知，根据信息准则，采取少数服从多数原则，取滞后阶数为3，此外取滞后阶数为2（SC为-6.182775）或取滞后阶数为3（SC为-6.105200）时，两者SC值相差不是很大。3.3模型参数

11、估计选取了最优滞后阶数3，进行模型的拟合。拟合结果如下：表4.5VAR（3）模型估计结果23由回归结果可知，VAR模型的参数估计一部分显着。估计的方程为：DLE=*DLE（-1）+*DLE（-2）+*DLE（-3）-*P（-1）-*P（-2）+*P（-3）+P=-*DLE（-1）+*DLE（-2）+*DLE（-3）+*P（-1）-*P（-2）-*P（-3）+4.4模型检验首先对模型进行平稳性检验表4.6VAR模型平稳性检验的表格显示RootsofCharacteristicPolynomialLagspecification:1327RootModulus0.883466-0.097039i0

12、.8887790.883466+0.097039i0.670300-0.321875-0.501863i0.596213-0.321875+0.501863i-0.2395920.239592Norootliesoutsidetheunitcircle.VARsatisfiesthestabilitycondition.图4.6VAR模型平稳性检验的图形显示由上表和上图可知，VAR模型的特征方程的根均在单位园内，因此VAR模型是平稳的。下面进行残差的自相关性的检验，检验结果如下：图4.7VAR模型各方程残差项的自相关图由上图可知，VAR模型允许不同方程的残差之间存在交叉相关性，但是残差自身不存

13、在自相关性，因此，观察残差自身的自相关图，可以看出自相关系数均位于置信区间内，说明残差不存在自相关性。第五部分模型应用5.1格兰杰因果检验接下来做两两变量之间的格兰杰因果检验。序列P与序列DLE:表5.1序列P与序列DLE格兰杰因果检验表PairwiseGrangerCausalityTests11/21/14Time:23:32Lags:ObsF-StatisticPdoesnotGrangerCauseDLE1082.777600.0451DLEdoesnotGrangerCauseP1.342860.2648由上述结果可知，在5%的置信水平下，P是dle的格兰杰原因，即全国房地产开发业综

14、合景气指数是人民币对美元汇率变动幅度的格兰杰原因。5.2脉冲响应由于脉冲响应函数收到变量顺序的影响，因此其结果与分析的主观因素有关，对于这三个变量：DLE、R、P,按照中国市场目前现状，认为DLE外生性最强，p其次最后为r。故选取顺序为DLE、P、R。图5.1脉冲响应图5.3方差分解表5.4方差分解结果VarianceDecompositionofDLE:PeriodS.E.100.00000.00371899.982500.0175000.00376998.893111.1068850.00392997.909522.0904810.00396696.365083.6349180.00401

15、994.218215.7817930.00407892.060357.9396490.00412989.8151510.184850.00418287.6054512.394550.00423185.5997514.400250.00427683.8063816.193620.00431682.2474817.752520.00435180.9365819.06342140.00438179.8538420.14616150.00440678.9772421.02276160.00442678.2831821.71682170.00444277.7439622.25604180.0044547

16、7.3334522.66655190.00446477.0279322.972070.00447176.8058323.19417210.00447676.6485523.351450.00447976.5405123.459490.00448276.4688723.53113240.00448376.4233623.57664250.00448476.3960023.60400260.00448576.3807123.6192976.3730623.62694280.00448676.3698923.630112976.3690623.630943076.3691923.630813176.3694723.6305376.3694623.630543376.3690023.631003476.3680723.631933576.3667523.633253676.3651623.63484