我国财政收入影响因素分析计量经济学论文要点.doc

-"我国财政收入影响因素分析"摘要：对我国财政收入影响因素进展了定量分析，建立了数学模型，并提出了提高我国财政收入质量的政策建议。关键词：财政收入 实证分析 影响因素一、 引言财政收入对于国民经济的运行及社会开展具有重要影响。首先，它是一个国家各项收入得以实现的物质保证。一个国家财政收入规模大小往往是衡量其经济实力的重要标志。其次，财政收入是国家对经济实行宏观调控的重要经济杠杆。宏观调控的首要问题是社会总需求与总供应的平衡问题，实现社会总需求与总供应的平衡，包括总量上的平衡和构造上的平衡两个层次的容。财政收入的杠杆既可通过增收和减收来发挥总量调控作用，也可通过对不同财政资金缴纳者的财政负担大小的调整，来发挥构造调整的作用。此外，财政收入分配也是调整国民收入初次分配格局，实现社会财富公平合理分配的主要工具。在我国，财政收入的主体是税收收入。因此，在税收体制及政策不变的情况下，财政收入会随着经济繁荣而增加，随着经济衰退而下降。我国的财政收入主要包括税收、国有经济收入、债务收入以及其他收入四种形式，因此，财政收入会受到不同因素的影响。从国民经济部门构造看，财政收入又表现为来自各经济部门的收入。财政收入的部门构成就是在财政收入中，由来自国民经济各部门的收入所占的不同比例来表现财政收入来源的构造，它表达国民经济各部门与财政收入的关系。我国财政收入主要来自于工业、农业、商业、交通运输和效劳业等部门。因此，本文认为财政收入主要受到总税收收入、国生产总值、其他收入和就业人口总数的影响。二、预设模型令财政收入Y亿元为被解释变量，总税收收入X1亿元、国生产总值X2亿元、其他收入X3亿元、就业人口总数为X4万人为解释变量，据此建立回归模型。二、 数据收集从"2010中国统计年鉴"得到1990-2009年每年的财政收入、总税收收入、国生产总值工、其他收入和就业人口总数的统计数据如下：obs财政收入Y总税收收入X1国生产总值X2其他收入X3就业人口总数X419902937.12821.8618667.8299.536474919913149.482990.1721781.5240.16549119923483.373296.9126923.5265.156615219934348.954255.335333.9191.046680819945218.15126.8848197.9280.186745519956242.26038.0460793.7396.196806519967407.996909.8271176.6724.666895019978651.148234.0478973682.36982019989875.959262.884402.3833.370637199911444.0810682.5889677.1925.4371394200013395.2312581.5199214.6944.9872085200116386.0415301.38109655.21218.173025200218903.6417636.45120332.71328.7473740200321715.2520017.31135822.81691.9374432200426396.4724165.68159878.32148.3275200200531649.2928778.54184937.42707.8375825200638760.234804.35216314.43683.8576400200751321.7845621.97265810.34457.9676990200861330.3554223.79314045.45552.4677480200968518.359521.59340506.97215.7277995三、 模型建立1、 散点图分析2、 单因素或多变量间关系分析YX1X2X3X4Y10.78530.08040.95640.1508X10.785310.84690.47820.3492X20.08040.846910.84710.0459X30.95640.47820.847110.0381X40.15080.34920.04590.03811由散点图分析和变量间关系分析可以看出被解释变量财政收入Y与解释变量总税收收入X1、国生产总值X2、其他收入X3、就业人口总数X4呈线性关系，因此该回归模型设为：3、 模型预模拟由eviews做ols回归得到结果：Dependent Variable: YMethod: Least SquaresDate: 11/14/11 Time: 17:51Sample: 1990 2009Included observations: 20VariableCoefficientStd. Errort-StatisticProb.  C7299.5231691.8144.3146140.0006X11.0628020.02110850.349720.0000X20.0017700.0045280.3910070.7013X30.8733690.1198067.2898520.0000X4-0.1159750.026580-4.3631600.0006R-squared0.999978    Mean dependent var20556.75Adjusted R-squared0.999972    S.D. dependent var19987.03S.E. of regression106.6264    Akaike info criterion12.38886Sum squared resid170537.9    Schwarz criterion12.63779Log likelihood-118.8886    F-statistic166897.9Durbin-Watson stat1.496517    Prob(F-statistic)0.000000 (4.314614) ( 50.34972 ) ( 0.391007) ( 7.289852) ( -4.363160)四、 模型检验1.计量经济学意义检验多重共线性检验与解决求相关系数矩阵，得到：Correlation MatrixYX1X2X3X410.78530.08040.95640.15080.785310.84690.47820.34920.08040.846910.84710.04590.95640.47820.847110.03810.15080.34920.04590.03811发现模型存在多重共线性。接下来运用逐步回归法对模型进展修正:将各个解释变量分别参加模型,进展一元回归:作Y与X1的回归，结果如下:Dependent Variable: YMethod: Least SquaresDate: 11/22/11 Time: 23:02Sample: 1990 2009Included observations: 20VariableCoefficientStd. Errort-StatisticProb.  C-755.6610145.2330-5.2030940.0001X11.1449940.005760198.79310.0000R-squared0.999545    Mean dependent var20556.75Adjusted R-squared0.999519    S.D. dependent var19987.03S.E. of regression438.1521    Akaike info criterion15.09765Sum squared resid3455590.    Schwarz criterion15.19722Log likelihood-148.9765    F-statistic39518.70Durbin-Watson stat0.475046    Prob(F-statistic)0.000000作Y与X2的回归，结果如下:Dependent Variable: YMethod: Least SquaresDate: 11/22/11 Time: 23:06Sample: 1990 2009Included observations: 20VariableCoefficientStd. Errort-StatisticProb.  C-5222.077861.2067-6.0636740.0000X20.2076890.00554837.432670.0000R-squared0.987317    Mean dependent var20556.75Adjusted R-squared0.986612    S.D. dependent var19987.03S.E. of regression2312.610    Akaike info criterion18.42478Sum squared resid96267005    Schwarz criterion18.52435Log likelihood-182.2478    F-statistic1401.205Durbin-Watson stat0.188013    Prob(F-statistic)0.000000作Y与X3的回归，结果如下:Dependent Variable: YMethod: Least SquaresDate: 11/22/11 Time: 23:08Sample: 1990 2009Included observations: 20VariableCoefficientStd. Errort-StatisticProb.  C2607.879773.99883.3693580.0034X310.030730.29431134.082090.0000R-squared0.984740    Mean dependent var20556.75Adjusted R-squared0.983893    S.D. dependent var19987.03S.E. of regression2536.645    Akaike info criterion18.60971Sum squared resid1.16E+08    Schwarz criterion18.70929Log likelihood-184.0971    F-statistic1161.589Durbin-Watson stat1.194389    Prob(F-statistic)0.000000作Y与X4的回归，结果如下:Dependent Variable: YMethod: Least SquaresDate: 11/22/11 Time: 23:08Sample: 1990 2009Included observations: 20VariableCoefficientStd. Errort-StatisticProb.  C-272959.337203.65-7.3368940.0000X44.0974030.5184677.9029180.0000R-squared0.776276    Mean dependent var20556.75Adjusted R-squared0.763846    S.D. dependent var19987.03S.E. of regression9712.824    Akaike info criterion21.29492Sum squared resid1.70E+09    Schwarz criterion21.39449Log likelihood-210.9492    F-statistic62.45611Durbin-Watson stat0.157356    Prob(F-statistic)0.000000依据可决系数最大的原那么选取X1作为进入回归模型的第一个解释变量,再依次将其余变量分别代入回归得:作Y与X1、X2的回归，结果如下Dependent Variable: YMethod: Least SquaresDate: 11/22/11 Time: 23:09Sample: 1990 2009Included observations: 20VariableCoefficientStd. Errort-StatisticProb.  C-188.4285239.0743-0.7881590.4415X11.2815940.04947225.905680.0000X2-0.0250550.009029-2.7749080.0130R-squared0.999687    Mean dependent var20556.75Adjusted R-squared0.999650    S.D. dependent var19987.03S.E. of regression374.0345    Akaike info criterion14.82405Sum squared resid2378330.    Schwarz criterion14.97341Log likelihood-145.2405    F-statistic27118.20Durbin-Watson stat0.683510    Prob(F-statistic)0.000000作Y与X1、X3的回归，结果如下Dependent Variable: YMethod: Least SquaresDate: 11/22/11 Time: 23:10Sample: 1990 2009Included observations: 20VariableCoefficientStd. Errort-StatisticProb.  C-351.105483.15053-4.2225270.0006X10.9928130.01870753.071960.0000X31.3569360.1651098.2184100.0000R-squared0.999908    Mean dependent var20556.75Adjusted R-squared0.999898    S.D. dependent var19987.03S.E. of regression202.1735    Akaike info criterion13.59361Sum squared resid694859.9    Schwarz criterion13.74297Log likelihood-132.9361    F-statistic92839.33Durbin-Watson stat1.177765    Prob(F-statistic)0.000000作Y与X1、X4的回归，结果如下Dependent Variable: YMethod: Least SquaresDate: 11/22/11 Time: 23:10Sample: 1990 2009Included observations: 20VariableCoefficientStd. Errort-StatisticProb.  C11853.461824.5226.4967480.0000X11.1858860.006645178.46080.0000X4-0.1866450.026984-6.9170030.0000R-squared0.999881    Mean dependent var20556.75Adjusted R-squared0.999867    S.D. dependent var19987.03S.E. of regression230.8464    Akaike info criterion13.85886Sum squared resid905931.0    Schwarz criterion14.00822Log likelihood-135.5886    F-statistic71206.90Durbin-Watson stat1.459938    Prob(F-statistic)0.000000在满足经济意义和可决系数的条件下选取X3作为进入模型的第二个解释变量,再次进展回归那么:作Y与X1、X3、X2的回归，结果如下Dependent Variable: YMethod: Least SquaresDate: 11/22/11 Time: 23:13Sample: 1990 2009Included observations: 20VariableCoefficientStd. Errort-StatisticProb.  C-76.04458100.1724-0.7591370.4588X11.0859240.02980136.438810.0000X31.2108530.1334449.0738770.0000X2-0.0140730.003944-3.5679010.0026R-squared0.999949    Mean dependent var20556.75Adjusted R-squared0.999939    S.D. dependent var19987.03S.E. of regression155.5183    Akaike info criterion13.10826Sum squared resid386975.0    Schwarz criterion13.30741Log likelihood-127.0826    F-statistic104602.9Durbin-Watson stat1.196933    Prob(F-statistic)0.000000作Y与X1、X3、X4的回归，结果如下Dependent Variable: YMethod: Least SquaresDate: 11/22/11 Time: 23:13Sample: 1990 2009Included observations: 20VariableCoefficientStd. Errort-StatisticProb.  C6781.7641024.7456.6180030.0000X11.0686420.01451473.627640.0000X30.8910690.1079498.2545510.0000X4-0.1076390.015451-6.9666750.0000R-squared0.999977    Mean dependent var20556.75Adjusted R-squared0.999973    S.D. dependent var19987.03S.E. of regression103.7654    Akaike info criterion12.29900Sum squared resid172276.1    Schwarz criterion12.49814Log likelihood-118.9900    F-statistic234970.9Durbin-Watson stat1.451447    Prob(F-statistic)0.000000可见参加其余任何一个变量都会导致系数符号与经济意义不符,故最终修正后的回归模型为:Dependent Variable: YMethod: Least SquaresDate: 11/30/11 Time: 12:18Sample: 1990 2009Included observations: 20VariableCoefficientStd. Errort-StatisticProb.  C-351.105483.15053-4.2225270.0006X10.9928130.01870753.071960.0000X31.3569360.1651098.2184100.0000R-squared0.999908    Mean dependent var20556.75Adjusted R-squared0.999898    S.D. dependent var19987.03S.E. of regression202.1735    Akaike info criterion13.59361Sum squared resid694859.9    Schwarz criterion13.74297Log likelihood-132.9361    F-statistic92839.33Durbin-Watson stat1.177765    Prob(F-statistic)0.000000(-4.222527) ( 53.07196) ( 8.218410)异方差检验与修正 图示法ee与X1的散点图如下：说明ee与X1存在单调递增型异方差性。ee与X3的散点图如下：说明ee与X3存在单调递增型异方差性。G-Q检验对20组数据剔除掉中间四组剩下的进展分组后，第一组1990-1997数据的回归结果：Dependent Variable: YMethod: Least SquaresDate: 11/30/11 Time: 12:54Sample: 1990 1997Included observations: 8VariableCoefficientStd. Errort-StatisticProb.  X10.9841230.01625560.543200.0000X30.8515180.1566885.4344720.0029C-28.3427545.36993-0.6247030.5596R-squared0.999686    Mean dependent var5179.791Adjusted R-squared0.999560    S.D. dependent var2099.840S.E. of regression44.05899    Akaike info criterion10.68893Sum squared resid9705.972    Schwarz criterion10.71872Log likelihood-39.75573    F-statistic7947.575Durbin-Watson stat1.663630    Prob(F-statistic)0.000000残差平方和RSS1=9705.972第二组2002-2009数据的回归结果：Dependent Variable: YMethod: Least SquaresDate: 11/30/11 Time: 12:55Sample: 2002 2009Included observations: 8VariableCoefficientStd. Errort-StatisticProb.  X11.0664040.02774738.433210.0000X30.8472280.2151143.9385030.0110C-1184.159261.8258-4.5226980.0063R-squared0.999932    Mean dependent var39824.41Adjusted R-squared0.999905    S.D. dependent var18639.16S.E. of regression182.0047    Akaike info criterion13.52594Sum squared resid165628.5    Schwarz criterion13.55573Log likelihood-51.10375    F-statistic36705.08Durbin-Watson stat1.326122    Prob(F-statistic)0.000000残差平方和RSS2= 165628.5所以F= RSS2/RSS1= 165628.5/9705.972=17.0646在给定a=5%下查得临界值 ，因此否认两组子样方差一样的假设，从而该总体随机项存在递增异方差性。White 方法检验White Heteroskedasticity Test:F-statistic6.142010    Probability0.003919Obs*R-squared12.41812    Probability0.014498Test Equation:Dependent Variable: RESID2Method: Least SquaresDate: 11/30/11 Time: 13:21Sample: 1990 2009Included observations: 20VariableCoefficientStd. Errort-StatisticProb.  C24856.5019211.301.2938480.2153X1-20.573277.549127-2.7252520.0156X120.0002128.04E-052.6399820.0186X3237.181378.613233.0170670.0087X32-0.0240730.006568-3.6652300.0023R-squared0.620906    Mean dependent var34743.00Adjusted R-squared0.519815    S.D. dependent var49156.00S.E. of regression34062.86    Akaike info criterion23.92212Sum squared resid1.74E+10    Schwarz criterion24.17105Log likelihood-234.2212    F-statistic6.142010Durbin-Watson stat1.560937    Prob(F-statistic)0.003919a=5%下，临界值拒绝同方差性修正Dependent Variable: YMethod: Least SquaresDate: 11/30/11 Time: 14:29Sample: 1990 2009Included observations: 20Weighting series: 1/E1VariableCoefficientStd. Errort-StatisticProb.  C-314.207443.68550-7.1924860.0000X10.9797580.008622113.63360.0000X31.4572910.06592222.106290.0000Weighted StatisticsR-squared0.999999    Mean dependent var27246.27Adjusted R-squared0.999999    S.D. dependent var74471.17S.E. of regression73.91795    Akai