影响国内私人汽车拥有量的几个重要因素分析.doc

影响国内私人汽车拥有量的几个重要因素分析内容摘要：本文主要是研究对我国私人汽车拥有量产生重要影响的几个因素。按照影响的重要程度，选择全国民用私人汽拥有量，全国人口数，全国居民消费水平指数，全国汽车产量，全国公路长度作为解释变量。模型建立后，利用EVIEWS软件对模型进行参数估计和检验，并加以校正。对最后的结果进行经济意义分析，然后提出自己的看法。关键词：全国民用私人汽车拥有量 全国人口数 全国居民消费水平指数 全国汽车年产量 全国公路长度 一 导论改革开放以来我国GDP增长速度比世界平均水平高出多倍，一直位居世界前列。随着综合国力的增强，人民生活水平大大提高。20世纪90年代以前，我国汽车市场处于公务车阶段，不仅需求量少，而且70来自政府、事业单位的公务用车，剩下的多是企业的商务用车，几乎没有什么私人用车。1990年至2000年，公务用车的份额下降，商务用车的份额加大，私人购车开始起步。2002年以来，私人购车占整个市场的份额迅速提升，进入私人购车阶段。今年，我们对汽车市场总体还是看好。汽车销量增长虽不会象去年那么快，但也可以达到20%左右。按照国际通用的车价和国内生产总值增长比较系数计算，未来10-15年中国有购车能力的人口可达5亿，约1.5亿个家庭。未来20年，中国有望成为全球第一大汽车市场。之所以这么有信心，是因为整个国民经济发展的势头仍然比较强劲。目前，汽车、住房、教育等行业列入了拉动内需的龙头行业。这都将使得我国今年的GDP持续高水平增长。在经济强劲发展的势头下，国民的收入水平也在提高，购买能力大大加强，同时扩大内需的政策极大地促进了中国汽车业的发展。汽车工业对国民经济的影响力越来越大。而目前来说，汽车特别是用于消费的私人轿车保有量的多少，与经济发展、经济活跃程度、国内生产总值、人均国内生产总值的增长，以及道路建设的发展，有着密切的联系。与此同时，消费者日益膨胀的购车热情和造汽车带来的暴利，引来了更多人的垂涎，一场新的汽车投资热风起云涌。在新一轮的造车运动中，上新车型几乎成为共同的选择，目标也均指向新的增长点私人轿车市场。同时随着居民消费结构的升级，私人购车呈现出迅猛增长的势头，成为我国汽车产业发展的决定性力量。目前，私人已经成为主要的购车群体。从私人汽车拥有量结构看，新增的私人汽车中近为小型和微型客车。通过对国际轿车市场研究发现，当车价与人均之比达到或时，是轿车进入家庭的转折点。目前，我国北京、上海、广州和深圳等大城市的车价与人均之比已经接近这个水平，私人购车进入了爆发性增长阶段。正因为私人汽车逐渐占据了汽车消费市场的主导地位，也直接反映了整个汽车行业的现状，私人汽车的消费市场成为了我们越来越关注的对象。为了实证对私人汽车消费市场的具体影响因素，以便于我们根据实证结果提出我们的政策建议，我们查找了一些关于我国汽车行业当今各方面的情况，选择用全国民用私人汽拥有量作为反映我国私人汽车消费市场现状的指标，并依据相关的数据资料选取了全国人口数（考虑到从某方面上反映了私人汽车市场的大小），全国居民消费水平指数（考虑到反映了我国居民对私人汽车市购买力程度），全国公路长度（也是对汽车市场消费造成影响的外部因素）作为解释变量。二 相关数据收集（数据来源：2002统计摘要 ）yx1x2x3x4时间全国民用私人汽全国人口全国居民消全国汽车产全国公路长车拥有量(万辆）数（万人）费水平指数量（万辆）度（万公里）198528.49105851323.543.7294.24198634.71107507337.836.9896.28198742.29109300356.947.1898.22198860.42111026383.464.4799.96198973.12112704381.458.35101.43199081.62114333394.451.4102.83199196.04115823427.171.42104.111992118.2117171482.3106.67105.671993155.77118517521.4129.85108.351994205.42119850543.8136.69111.781995249.96121121584.6145.27115.71996289.67122389637.8147.52118.581997358.36123626664.4158.25122.641998423.65124810701.3163127.851999512.35125836754.2174.3130.462000704.9126987786.1185.6139.162001821.64128036801.3209.5145.892002945.7130549843.1236.76152.7原始数据的分析记录在附录1由于原模型存在自相关，我们决定通过改变解释变量模式从而消除由于样本数量不足带来的问题。新模型中：Y/X1Y 人均民用汽车私人拥有量X3/X1X1人均汽车产量X2X2 全国居民消费指数X4/X1X3 人均公路长度时间y/x1x3/x1x4/x119850.0002690.0004130.0008919860.0003230.0003440.00089619870.0003870.0004320.00089919880.0005440.0005810.000919890.0006490.0005180.000919900.0007140.000450.00089919910.0008290.0006170.00089919920.0010090.000910.00090219930.0013140.0010960.00091419940.0017140.0011410.00093319950.0020640.0011990.00095519960.0023670.0012050.00096919970.0028990.001280.00099219980.0033940.0013060.00102419990.0040720.0013850.00103720000.0055510.0014620.00109620010.0064170.0016360.00113920020.0072440.0018140.00117三 平稳性检验Augmented dickey_fuller unit root test on D (Y1，2)ADF Test Statistic-4.437656 1% Critical Value*-4.9893 5% Critical Value-3.8730 10% Critical Value-3.3820*MacKinnon critical values for rejection of hypothesis of a unit root.Augmented Dickey-Fuller Test EquationDependent Variable: D(Y,3)Method: Least SquaresDate: 06/15/05 Time: 20:08Sample(adjusted): 1991 2002Included observations: 12 after adjusting endpointsVariableCoefficientStd. Errort-StatisticProb. D(Y(-1),2)-4.3758780.986079-4.4376560.0044D(Y(-1),3)2.9263250.9167543.1920510.0188D(Y(-2),3)1.5436610.8190691.8846530.1085D(Y(-3),3)2.2029800.5752423.8296600.0087C-0.0002910.000166-1.7485570.1310TREND(1985)5.07E-051.75E-052.9026000.0272R-squared0.963765 Mean dependent var-4.17E-09Adjusted R-squared0.933569 S.D. dependent var0.000517S.E. of regression0.000133 Akaike info criterion-14.70026Sum squared resid1.07E-07 Schwarz criterion-14.45781Log likelihood94.20157 F-statistic31.91688Durbin-Watson stat2.451776 Prob(F-statistic)0.000300Augmented dickey_fuller unit root test on D (x1,1)ADF Test Statistic-3.492868 1% Critical Value*-4.8025 5% Critical Value-3.7921 10% Critical Value-3.3393*MacKinnon critical values for rejection of hypothesis of a unit root.Augmented Dickey-Fuller Test EquationDependent Variable: D(X1)Method: Least SquaresDate: 06/15/05 Time: 20:18Sample(adjusted): 1989 2002Included observations: 14 after adjusting endpointsVariableCoefficientStd. Errort-StatisticProb. X1(-1)-0.9929380.284276-3.4928680.0082D(X1(-1)1.0610820.2565354.1362170.0033D(X1(-2)-0.0393390.273775-0.1436910.8893D(X1(-3)0.4459430.2287311.9496380.0870C7.54E-055.07E-051.4860210.1756TREND(1985)8.97E-052.50E-053.5950670.0070R-squared0.779700 Mean dependent var8.81E-05Adjusted R-squared0.642013 S.D. dependent var0.000102S.E. of regression6.08E-05 Akaike info criterion-16.27936Sum squared resid2.96E-08 Schwarz criterion-16.00547Log likelihood119.9555 F-statistic5.662833Durbin-Watson stat2.501918 Prob(F-statistic)0.015901 Augmented dickey_fuller unit root test on D (x2,2)ADF Test Statistic-3.882038 1% Critical Value*-4.7315 5% Critical Value-3.7611 10% Critical Value-3.3228*MacKinnon critical values for rejection of hypothesis of a unit root.Augmented Dickey-Fuller Test EquationDependent Variable: D(X2,3)Method: Least SquaresDate: 06/14/05 Time: 10:56Sample(adjusted): 1988 2002Included observations: 15 after adjusting endpointsVariableCoefficientStd. Errort-StatisticProb. D(X2(-1),2)-1.1995810.309008-3.8820380.0022C3.89526513.935620.2795190.7846TREND(1985)-0.2369961.283314-0.1846750.8566R-squared0.563244 Mean dependent var1.453333Adjusted R-squared0.490452 S.D. dependent var29.42530S.E. of regression21.00456 Akaike info criterion9.104213Sum squared resid5294.301 Schwarz criterion9.245823Log likelihood-65.28160 F-statistic7.737652Durbin-Watson stat2.152035 Prob(F-statistic)0.006941Augmented dickey_fuller unit root test on D (x3)ADF Test Statistic-4.293548 1% Critical Value*-4.6712 5% Critical Value-3.7347 10% Critical Value-3.3086*MacKinnon critical values for rejection of hypothesis of a unit root.Augmented Dickey-Fuller Test EquationDependent Variable: D(X3,2)Method: Least SquaresDate: 06/15/05 Time: 20:23Sample(adjusted): 1987 2002Included observations: 16 after adjusting endpointsVariableCoefficientStd. Errort-StatisticProb. D(X3(-1)-1.1510160.268080-4.2935480.0009C-1.36E-056.83E-06-1.9868610.0684TREND(1985)3.48E-069.74E-073.5728920.0034R-squared0.586622 Mean dependent var1.56E-06Adjusted R-squared0.523025 S.D. dependent var1.50E-05S.E. of regression1.04E-05 Akaike info criterion-19.94795Sum squared resid1.40E-09 Schwarz criterion-19.80309Log likelihood162.5836 F-statistic9.224093Durbin-Watson stat2.128125 Prob(F-statistic)0.003208可以看出，解释变量X1在0阶差分，X2在2阶差分，X3在1阶差分，Y在2阶差分情况下通过了平稳性检验。四 Granger Causality检验Pairwise Granger Causality TestsDate: 06/15/05 Time: 20:26Sample: 1985 2002 Lags: 1 Null Hypothesis:ObsF-StatisticProbability X1 does not Granger Cause Y17 1.27559 0.27770 Y does not Granger Cause X1 1.34361 0.26579滞后长度1 结论：X1Y 拒绝 Y X1 拒绝Pairwise Granger Causality TestsDate: 06/15/05 Time: 20:27Sample: 1985 2002 Lags: 1 Null Hypothesis:ObsF-StatisticProbability X2 does not Granger Cause Y17 5.20861 0.03863 Y does not Granger Cause X2 1.54629 0.23411滞后长度1 结论：X2Y 接受 Y X2 拒绝Pairwise Granger Causality TestsDate: 06/15/05 Time: 20:29Sample: 1985 2002 Lags: 2 Null Hypothesis:ObsF-StatisticProbability X3 does not Granger Cause Y16 4.39300 0.03960 Y does not Granger Cause X3 5.59132 0.02111滞后长度2 结论：X3Y 接受 Y X3 接受五 参数的估计Dependent Variable: YMethod: Least SquaresDate: 06/15/05 Time: 20:25Sample: 1985 2002Included observations: 18VariableCoefficientStd. Errort-StatisticProb. C-0.0193820.000828-23.416700.0000X10.6329890.3954581.6006470.1318X2-3.54E-071.33E-06-0.2655860.7944X321.988031.21339618.121060.0000R-squared0.995934 Mean dependent var0.002320Adjusted R-squared0.995063 S.D. dependent var0.002195S.E. of regression0.000154 Akaike info criterion-14.52355Sum squared resid3.33E-07 Schwarz criterion-14.32569Log likelihood134.7120 F-statistic1143.175Durbin-Watson stat0.997884 Prob(F-statistic)0.000000得Y=-0.019382+0.632989X1-3.54E-07X2+21.98803X3 T （-23.41670）（1.600647）（-0.265586）（18.12106） 可决系数R2=0.995934 修正值为0.995063 从拟合图形看出（特别是残差图）做出的模型拟合效果不是非常理想，所以决定通过检验来进一步修正该模型。六 模型的检验和修正1. 经济意义检验 首先在经济意义上通不过。X2的系数为负，全国居民消费水平指数的提高，全国私人汽拥有量反而在下降。这显然与经济事实不符。2统计推断检验 从回归结果来看，模型的整体拟合度较高（可决系数R2=0.995934）,F统计量的值在给定显著性水平=0.05的情况下很显著（F=1143.175），但是C，X1，X2，的T统计值均不显著（它们的T统计量的值的绝对值均小于2），说明这些变量对Y的影响不显著，或者变量之间可能存在多重共线性的影响使其T值不显著。3计量经济学检验（1）多重共线性检验X1和Y回归Dependent Variable: YMethod: Least SquaresDate: 06/26/05 Time: 16:00Sample: 1985 2002Included observations: 18VariableCoefficientStd. Errort-StatisticProb. C-0.0041920.000581-7.2157310.0000X11.18E-051.01E-0611.735610.0000R-squared0.895918 Mean dependent var0.002320Adjusted R-squared0.889413 S.D. dependent var0.002195S.E. of regression0.000730 Akaike info criterion-11.50317Sum squared resid8.52E-06 Schwarz criterion-11.40424Log likelihood105.5285 F-statistic137.7245Durbin-Watson stat0.281816 Prob(F-statistic)0.000000X2和Y回归Dependent Variable: YMethod: Least SquaresDate: 06/26/05 Time: 16:00Sample: 1985 2002Included observations: 18VariableCoefficientStd. Errort-StatisticProb. C-0.0019280.000531-3.6298120.0023X24.2983120.4888108.7934150.0000R-squared0.828555 Mean dependent var0.002320Adjusted R-squared0.817839 S.D. dependent var0.002195S.E. of regression0.000937 Akaike info criterion-11.00409Sum squared resid1.40E-05 Schwarz criterion-10.90516Log likelihood101.0368 F-statistic77.32414Durbin-Watson stat0.304057 Prob(F-statistic)0.000000X3和Y回归Dependent Variable: YMethod: Least SquaresDate: 06/26/05 Time: 16:01Sample: 1985 2002Included observations: 18VariableCoefficientStd. Errort-StatisticProb. C-0.0211340.000485-43.570170.0000X324.243530.49933948.551210.0000R-squared0.993258 Mean dependent var0.002320Adjusted R-squared0.992837 S.D. dependent var0.002195S.E. of regression0.000186 Akaike info criterion-14.24001Sum squared resid5.52E-07 Schwarz criterion-14.14108Log likelihood130.1601 F-statistic2357.220Durbin-Watson stat0.650295 Prob(F-statistic)0.000000因为X3和Y回归结果较理想，加入X1X3,X1Dependent Variable: YMethod: Least SquaresDate: 06/26/05 Time: 16:01Sample: 1985 2002Included observations: 18VariableCoefficientStd. Errort-StatisticProb. C-0.0192530.000860-22.385480.0000X321.412871.20912017.709470.0000X11.56E-066.20E-072.5073070.0242R-squared0.995249 Mean dependent var0.002320Adjusted R-squared0.994616 S.D. dependent var0.002195S.E. of regression0.000161 Akaike info criterion-14.47893Sum squared resid3.89E-07 Schwarz criterion-14.33053Log likelihood133.3103 F-statistic1571.180Durbin-Watson stat0.740153 Prob(F-statistic)0.000000X3，X2和Y回归Dependent Variable: YMethod: Least SquaresDate: 06/26/05 Time: 16:02Sample: 1985 2002Included observations: 18VariableCoefficientStd. Errort-StatisticProb. C-0.0192770.000699-27.588250.0000X321.772340.87053925.010200.0000X20.5393750.1689903.1917580.0061R-squared0.995985 Mean dependent var0.002320Adjusted R-squared0.995450 S.D. dependent var0.002195S.E. of regression0.000148 Akaike info criterion-14.64719Sum squared resid3.29E-07 Schwarz criterion-14.49880Log likelihood134.8247 F-statistic1860.470Durbin-Watson stat0.908252 Prob(F-statistic)0.000000观察结果，我们最终剔除变量X1，保留X2和X3得到模型 Y=-0.019277+21.77234X3+0.539375X2（2）自相关检验检验：从模型设定来看，没有违背D-W检验的假设条件，因此可以用D-W检验来检验模型是否存在自相关。 根据上表中估计的结果，由DW=1.060181，给定显著性水平=0.05，查Durbin-Watson表，n=18,k=3,得dl=0.933,du=1.696。因为DW统计量在dl与du之间（0.9331.0601811.696），根据判定区域，不能判定是否存在一阶自相关，这可能是样本容量太小而造成的。所以在这里暂时不对模型进行自相关的修正。3）异方差检验 检验：由于样本不大，采用的是时间序列数据，所以决定用ARCH检验法检验模型是否存在异方差。结果如下：（滞后期数为3,即p=3n=18）ARCH Test:F-statistic0.295864 Probability0.827661Obs*R-squared1.119983 Probability0.772252Test Equation:Dependent Variable: RESID2Method: Least SquaresDate: 06/15/05 Time: 21:21Sample(adjusted): 1988 2002Included observations: 15 after adjusting endpointsVariableCoefficientStd. Errort-StatisticProb. C2.88E-081.43E-082.0161570.0689RESID2(-1)-0.1644860.300911-0.5466270.5955RESID2(-2)-0.0362020.307793-0.1176190.9085RESID2(-3)-0.2238690.301902-0.7415280.4739R-squared0.074666 Mean dependent var2.01E-08Adjusted R-squared-0.177698 S.D. dependent var2.72E-08S.E. of regression2.95E-08 Akaike info criterion-31.61405Sum squared resid9.60E-15 Schwarz criterion-31.42524Log likelihood241.1054 F-statistic0.295864Durbin-Watson stat2.094508 Prob(F-statistic)0.827661 Probability0.827661 在H0：误差项不存在异方差 的假设下，犯拒真错误的概率是0.827661。这是一个很大的概率。于是我们接受H0六 对模型的经济意义解释及存在的问题（以及政策建议）从对被解释变量及各解释变量的时间序列的分析，可以得出人均私用车拥有量Y为2阶单整，消费指数X2为2阶单整，人均公路里程为单位根序列，对三个时间序列作简单回归，得到Y的最终模型为：Y=-0.019277+21.77234X3+0.539375X2 通过该回归模型可以有以下结论：（1）我国人均私家车的拥有量主要受居民消费水平指数及人均公路里程数的影响，而且均存在正向相关关系；（2）通过因果检验表明，人均私用车拥有量与人均公路里程存在互为因果的关系，这说明随着我国公路建设加强，使居民对私家车的需求上升，同时由于私家车增多对交通产生的压力也促使我国公路的建设和改造；（3）从相关系数来看，公路建设水平对私家