多重共线性的检验和克服

多重共线性的检验和克服一、实验目的掌握多重共线性的检验和处理方法二、实验原理判定系数检验法、解释变量相关系数矩阵法、逐步回归法三、实验步骤（一）创建新工作文件打开EViews，执行以下步骤：点击Filenewworkfile，然后分别输入起始时间和结束时间（1983和2000），最后保存。（二）输入数据表1中国粮食生产函数模型年份yx1x2x3x4x51983387281659.811404716209.31802231645.11984407311739.81128841526419497316851985379111775.810884522705.32091330351.51986391511930.61109332365622950304671987402081999.311126820392.724836308701988394082141.511012323944.72657531455.71989407552357.111220524448.72806732440.51990446242590.311346617819.32870833330.41991435292806.1112314278142938934186.31992442642930.211056025894.730308340371993456493151.9110509231333181733258.21994445103317.9109544313833380232690.31995466623593.7110060222673611832334.51996504543827.9112548212333854732260.41997494173980.7112912303094201632434.91998512304083.7113787251814520832626.41999508394124.3113161267314899632911.82000462184146.4108463343745257432797.5其中：Y表示粮食产量，X1表示农业化肥施用量，X2表示粮食播种面积，X3表示成灾面积，X4表示农业机械总动力，X5表示农业劳动力。点击Quickemptygroup，然后输入变量、导入数据，点击name保存。（三）普通最小二乘法估计模型参数点击estimateequation，输入变量y、c、x1、x2、x3、x4、x5，运用普通最小二乘法进行回归分析，结果如表2。DependentVariable:YMethod:LeastSquaresDate:12/22/10Time:10:09Sample:19832000Includedobservations:18VariableCoefficientStd.Errort-StatisticProb.C-12815.7514078.90-0.9102800.3806X16.2125620.7408818.3853730.0000X20.4213800.1269253.3199190.0061X3-0.1662600.059229-2.8070650.0158X4-0.0977700.067647-1.4452990.1740X5-0.0284250.202357-0.1404710.8906R-squared0.982798Meandependentvar44127.11AdjustedR-squared0.975630S.D.dependentvar4409.100S.E.ofregression688.2984Akaikeinfocriterion16.16752Sumsquaredresid5685056.Schwarzcriterion16.46431Loglikelihood-139.5077F-statistic137.1164Durbin-Watsonstat1.810512Prob(F-statistic)0.000000得到的估计模型如下（依次点击view、representations）：Y=-12815.75054+6.21256194*X1+0.4213801204*X2-0.1662595096*X3-0.09776959861*X4-0.02842514979*X5（四）多重共线性检验由表2可知，x1、x2系数的t检验值的绝对值小于2，不显著，可能存在多重共线性。计算各解释变量的相关系数，得相关关系矩阵，结果见表3：YX1X2X3X4X5Y10.9444260265550.273994735710.3994537033250.8675870897580.553560360594X10.94442602655510.01182347844880.6401749943390.9602777864530.545450492387X20.273994735710.01182347844881-0.454908412062-0.03847935731220.182359187794X30.3994537033250.640174994339-0.45490841206210.689565016210.355735263453X40.8675870897580.960277786453-0.03847935731220.6895650162110.45416886948X50.5535603605940.5454504923870.1823591877940.3557352634530.454168869481由相关关系矩阵可以看出，x1与x4之间可能存在高度相关。（五）逐步回归法被解释变量y分别对解释变量进行一元回归（点击quickestimateequation）,得知x1的方程2R最大，以x1为基础，顺次加入其他变量逐步回归。经回归结果分析，剔除x4、x5两个变量，得到最终分析结果如表3：DependentVariable:YMethod:LeastSquaresDate:12/22/10Time:10:52Sample:19832000Includedobservations:18VariableCoefficientStd.Errort-StatisticProb.C-11978.1814072.92-0.8511510.4090X15.2559350.26859519.568280.0000X20.4084320.1219743.3485220.0048X3-0.1946090.054533-3.5686370.0031R-squared0.979593Meandependentvar44127.11AdjustedR-squared0.975220S.D.dependentvar4409.100S.E.ofregression694.0715Akaikeinfocriterion16.11616Sumsquaredresid6744293.Schwarzcriterion16.31402Loglikelihood-141.0454F-statistic224.0086Durbin-Watsonstat1.528658Prob(F-statistic)0.000000最后修正多重共线性影响的回归结果为：Y=-11978.18057+5.255935121*X1+0.408432175*X2-0.1946087795*X3^