实验三多元线性回归模型的估计和检验

实验报告课程名称：计量经济学实验项目：实验三多元线性回归模型的估计和检验实验类型：综合性□设计性□验证性专业班别：姓名：学号：实验课室：厚德楼B503指导教师：石立实验日期：2016年4月29日广东商学院华商学院教务处制一、实验项目训练方案小组合作：是□否小组成员：无实验目的：掌握多元线性回归模型估计和检验的方法。实验场地及仪器、设备和材料实验室：普通配置的计算机，Eviews软件及常用办公软件。实验训练内容（包括实验原理和操作步骤）：【实验步骤】（一）国内生产总值的增长模型：分析广东省国内生产总值的增长，根据广东数据（数据见“表：广东省宏观经济数据-第三章.xls”文件，各变量的表示按照试验指导课本上的来表示）选择不变价GDP（GDPB）、不变价资本存量（ZC）和从业人员（RY），把GDPB作为因变量，ZC和RY作为两个解释变量进行二元线性回归分析。要求：按照试验指导课本100P~102P，分别作：1．作散点图（GDPB同ZC，GDPB同RY）（结果控制在本页）01,0002,0003,0004,0005,0006,0007,00004,0008,00012,00016,000ZCGDPB01,0002,0003,0004,0005,0006,0007,0002,0003,0004,0005,0006,000RYGDPB2．进行因果关系检验（GDPB同ZC，GDPB同RY）（结果控制在本页）PairwiseGrangerCausalityTestsDate:04/29/16Time:14:35Sample:19782005Lags:2NullHypothesis:ObsF-StatisticProb.ZCdoesnotGrangerCauseGDPB263.849390.0376GDPBdoesnotGrangerCauseZC19.07482.E-05PairwiseGrangerCausalityTestsDate:04/29/16Time:14:36Sample:19782005Lags:2NullHypothesis:ObsF-StatisticProb.RYdoesnotGrangerCauseGDPB260.096030.9088GDPBdoesnotGrangerCauseRY4.728560.02023．作GDPB同ZC和RY的多元线性回归，写出模型估计的结果，并分析模型检验是均否通过？（三个检验）（结果控制在本页）DependentVariable:GDPBMethod:LeastSquaresDate:04/29/16Time:14:40Sample:19782005Includedobservations:28CoefficientStd.Errort-StatisticProb.ZC0.3771700.00835545.142650.0000RY0.3536890.0427578.2720280.0000C-800.5997113.7822-7.0362470.0000R-squared0.999152Meandependentvar1754.112AdjustedR-squared0.999085S.D.dependentvar1683.912S.E.ofregression50.94570Akaikeinfocriterion10.80035Sumsquaredresid64886.61Schwarzcriterion10.94309Loglikelihood-148.2050Hannan-Quinncriter.10.84399F-statistic14736.32Durbin-Watsonstat0.443892Prob(F-statistic)0.000000得到估计方程GDPB=0.37716969694502*ZC+0.353688537498*RY--800.599732335估计方程的判定系数R2接近1；参数显著性t检验值均大于2；方程显著性F检验显著。调整的判定系数为0.999085，比下面的一元回归有明显改善。4．将建立的二元回归模型（GDPB同ZC和RY）同一元回归模型（GDPB同ZC、GDPB同RY）相比较，分析优点。（结果控制在本页）一元回归模型：DependentVariable:GDBMethod:LeastSquaresDate:04/29/16Time:14:52Sample:19782005Includedobservations:28CoefficientStd.Errort-StatisticProb.ZC0.4428980.00489690.460000.0000C133.972125.570545.2393140.0000R-squared0.996833Meandependentvar1754.112AdjustedR-squared0.996711S.D.dependentvar1683.912S.E.ofregression96.57302Akaikeinfocriterion12.04722Sumsquaredresid242485.0Schwarzcriterion12.14238Loglikelihood-166.6611Hannan-Quinncriter.12.07632F-statistic8183.011Durbin-Watsonstat0.167556Prob(F-statistic)0.000000DependentVariable:GDPBMethod:LeastSquaresDate:04/29/16Time:14:54Sample:19782005Includedobservations:28CoefficientStd.Errort-StatisticProb.RY2.1893170.11773518.595280.0000C-5519.137400.4253-13.783190.0000R-squared0.930067Meandependentvar1754.112AdjustedR-squared0.927377S.D.dependentvar1683.912S.E.ofregression453.7907Akaikeinfocriterion15.14190Sumsquaredresid5354077.Schwarzcriterion15.23706Loglikelihood-209.9866Hannan-Quinncriter.15.17099F-statistic345.7844Durbin-Watsonstat0.078643Prob(F-statistic)0.000000问题3的二元回归模型与一元回归模型比较，可以得出估计方程的判定系数R2、参数显著性t检验、方程显著性F检验和调整的判定系数有些比一元回归有改进，表明这些确实应该进行二元回归。5．结合相关的经济理论，分析估计的二元回归模型的经济意义。（结果控制在本页）因为GDPB=1.669146*ZC-0.057889*RY-476.1072系数说明，不变价资本存量ZC每增加1.669146个单位，同时从业人员RY0.057889个单位，国内生产总值GDPS增加1个单位，因此符合经济理论。（二）宏观经济模型：根据广东数据，研究广东省居民消费行为、固定资产投资行为、货物和服务净出口行为和存货行为，分别建立居民消费模型、固定资产投资模型、货物和服务净出口模型和存货增加模型。要求：按照试验指导课本510P~211P，分别作出以下模型，并对需要改进的模型进行改进。写出最终估计的模型结果，并结合相关的经济理论，分析模型的经济意义。（数据见“表：广东省宏观经济数据-第三章.xls”文件，各变量的表示按照试验指导课本上的来表示。）1．居民消费模型（结果控制在本页）02,0004,0006,0008,00010,00002,0004,0006,0008,00010,000LBXFJ02,0004,0006,0008,00010,00002,0004,0006,0008,00010,000LBXFJPairwiseGrangerCausalityTestsDate:04/29/16Time:15:15Sample:19782005Lags:2NullHypothesis:ObsF-StatisticProb.LBdoesnotGrangerCauseXFJ267.190100.0042XFJdoesnotGragerCauseLB5.455160.0124DependentVariable:XFJMethod:LeastSquaresDate:04/29/16Time:15:17Sample:19782005Includedobservations:28VariableCoefficientStd.Errort-StatisticProb.LB0.7408080.03289322.521990.0000YY0.3620750.0464527.7946920.0000C46.9151336.602821.2817350.2117R-squared0.997789Meandependentvar2362.277AdjustedR-squared0.997612S.D.dependentvar2565.722S.E.ofregression125.3710Akaikeinfocriterion12.60139Sumsquaredresid392946.9Schwarzcriterion12.74412Loglikelihood-173.4194Hannan-Quinncriter.12.64502F-statistic5641.541Durbin-Watsonstat1.122075Prob(F-statistic)0.0000002．固定资产投资模型（结果控制在本页）DependentVariable:TZGMethod:LeastSquaresDate:04/29/16Time:15:21Sample:19782005Includedobservations:28VariableCoefficientStd.Errort-StatisticProb.ZJ1.1118640.2431524.527160.0001YY0.4316920.0525668.2123520.0000CZ0.1432100.4053080.3533380.7269C31.2762527.825171.1240270.2721R-squared0.997573Meandependentvar1628.997AdjustedR-squared0.997270S.D.dependentvar2003.852S.E.ofregression104.7010Akaikeinfocriterion12.27166Sumsquaredresid263095.1Schwarzcriterion12.46197Loglikelihood-167.8032Hannan-Quinncriter.12.32984F-statistic3288.646Durbin-Watsonstat1.298515Prob(F-statistic)0.0000003．货物和服务净流出模型（结果控制在本页）DependentVariable:CKMethod:LeastSquaresDate:04/29/16Time:15:25Sample:19782005Includedobservations:28VariableCoefficientStd.Errort-StatisticProb.GDP0.0882390.00552515.970670.0000LL-42.6598911.83064-3.6058800.0014C202.217395.250382.1230080.0438R-squared0.950564Mean