Eviews试验报告材料

1、实用文档江西农业大学经济贸易学院学生实验报告课程名称：计量经济学专业班级：经济1201班姓 名：学 号：指导教师：徐冬梅职 称：讲师实验日期：2021.12.11学生实验报告学生姓名学号组员：实验工程EVIEWS勺使用已必修 选修 冈演示性实验 団验证性实验 操作性实验 综合性实验实验地点治理模拟实验室实验仪器台号指导教师实验日期及节次一、实验目的及要求1、目的会使用EVIEWS寸计量经济模型进行分析2、内容及要求1对经典线形回归模型进行参数估计、参数的检验与区间估计, 对模型总体进行显著性检验；2异方差的检验及其处理；3自相关的检验及其处理；4多重共线性检验及其处理；二、仪器用具仪器名称规格

2、/型号数量备注计算机1无网络环境Eviews1三、实验方法与步骤一数据的输入、描述及其图形处理；二方程的估计；三参数的检验、违背经典假定的检验;四模型的处理与预测四、实验结果与数据处理实验一：中国城镇居民人均消费支出模型数据散点图：25000 20000 - 】*Y 15000 -忙 *»10000 - 八匚5000 -I11116000800010000120001400016000X通过Eviews估计参数方程回归方程：Depe ndent Variable: YMethod: Least SquaresDate: 11/2 7/14 Time: 15:02Sample: 1 3

3、1In cluded observati ons: 31VariableCoefficie ntStd. Error t-StatisticProb.X1.3594770.04330231.395250.0000C-57.90655377.7595-0.1532890.8792R-squared0.971419Mean depe ndent var11363.69Adjusted R-squared0.970433S.D.dependent var3294.469S.E. of regressi on566.4812Akaike info15.57911criteri onSum square

4、d resid9306127.Schwarz criteri on15.67162Log likelihood-239.4761F-statistic985.6616Durb in-Watson stat1.294974Prob(F-statistic)0.000000得出估计方程为：丫 = 1.35947661442*X - 57.9065479515异方差检验1、图示检验法15000001000000 -E 500000 - Bi0 _6000800010000120001400016000图形呈现离散趋势,大致判断存在异方差性.2、Park检验Depe ndent Variable: L

5、OG(E2)Method: Least SquaresDate: 11/2 7/14 Time: 16:16Sample: 1 31In cluded observati ons: 31VariableCoefficie ntStd. Error t-StatisticProb.C19.8256219.853590.9985910.3263LOG(X)-0.9564032.204080-0.4339240.6676R-squared0.006451Mean depe ndent var11.21371Adjusted R-squared-0.027809S.D.dependent var2.8

6、94595S.E. of regressi on2.934568Akaike info5.053338criteri onSum squared resid249.7389Schwarz criteri on5.145854Log likelihood-76.32674F-statistic0.188290Durb in-Watson stat2.456500Prob(F-statistic)0.667555看到图中L0G(E2冲P值为0.6676 > 0.05,所以不存在异方差性3、G-Q检验ei检验:Depe ndent Variable: XMethod: Least Square

7、sDate: 11/2 7/14 Time: 16:41Sample: 1 12In cluded observati ons: 12VariableCoefficie ntStd. Error t-StatisticProb.C4642.0282021.1832.3046710.0439Y0.2310460.2158241.0705300.3095R-squared0.102820Mean depe ndent var6796.390Adjusted R-squared0.013102S.D.dependent var293.2762S.E. of regressi on291.3486Ak

8、aike info14.33793criteri onSum squared resid848840.2Schwarz criteri on14.41875Log likelihood-84.02758F-statistic1.146034Durb in-Watson stat0.445146Prob(F-statistic)0.309538e2检验:Depe ndent Variable: XMethod: Least SquaresDate: 11/2 7/14 Time: 16:42Sample: 20 31In cluded observati ons: 12VariableCoeff

9、icie ntStd. Error t-StatisticProb.C583.4526593.43700.9831750.3487Y0.6977480.04019617.358700.0000R-squared0.967879Mean depe ndent var10586.89Adjusted R-squared0.964667S.D.dependent var2610.864S.E. of regressi on490.7655Akaike info15.38082criteri onSum squared resid2408507.Schwarz criteri on15.46164Lo

10、g likelihood-90.28493F-statistic301.3245Durb in-Watson stat2.748144Prob(F-statistic)0.000000第一个图中的残差平方和为 848840.2第二个图中的残差平方和为 2408507,所以不存在异方差性所以 F 值为 2408507/848840.2 = 2.8374 < 2.974、White 检验White Heteroskedasticity Test:F-statisticObs*R-squared2.240402Probability4.276524Probability0.1251520.11

11、7860Test Equati on:Depe ndent Variable: RESIDEMethod: Least SquaresDate: 11/2 7/14 Time: 16:50Sample: 1 31In eluded observati ons: 31VariableCoefficie ntStd. Error t-StatisticProb.C-2135113.1158576.-1.8428760.0760X503.7331242.20782.0797560.0468XA2-0.0236090.011650-2.0265900.0523R-squared0.137952Mean

12、 depe ndent var300197.6Adjusted R-squared0.076378S.D.dependent var347663.4S.E. of regressi on334122.9Akaike info28.36817criteri onSum squared resid3.13E+12Schwarz criteri on28.50694Log likelihood-436.7067F-statistic2.240402Durb in-Watson stat1.871252Prob(F-statistic)0.125152P值为0.11786 > 0.05,所以不存

13、在异方差性通过四种不同的检验得知除了图示检验法得出异方差的结论,其他的检验的结论都是不存在异方差的.5、WLS加权最小二乘法修正Depe ndent Variable: YMethod: Least SquaresDate: 11/2 7/14 Time: 17:14Sample: 1 31In cluded observati ons: 31Weight ing series: E3VariableCoefficie ntStd. Error t-StatisticProb.C-85.6942624.15675-3.5474250.0013X1.3622210.002307590.56150

14、.0000Weighted StatisticsR-squared1.000000Mean depe ndent var13474.53Adjusted R-squared1.000000S.D.dependent var61353.74S.E. of regressi on27.93264Akaike info9.559810criteri onSum squared resid22626.73Schwarz criteri on9.652325Log likelihood-146.1770F-statistic348762.9Durb in-Watson stat2.061818Prob(

15、F-statistic)0.000000Un weightedStatisticsR-squared0.971413Mean depe ndent var11363.69Adjusted R-squared0.970427S.D.dependent var3294.469S.E. of regressi on566.5415Sum squared resid9308110.Durb in-Watson stat2.178992实验二：中国粮食生产函数1、回归方程Depe ndent Variable: LOG(Y)Method: Least SquaresDate: 12/11/14 Time

16、: 15:06Sample: 1983 2007In cluded observati ons: 25VariableCoefficie ntStd. Error t-StatisticProb.LOG(X1)0.3811450.0502427.5861820.0000LOG(X2)1.2222890.1351799.0420300.0000LOG(X3)-0.0811100.015304-5.3000240.0000LOG(X4)-0.0472290.044767-1.0549800.3047LOG(X5)-0.1011740.057687-1.7538530.0956C-4.1731741

17、.923624-2.1694340.0429R-squared0.981597Mean depe ndent var10.70905Adjusted R-squared0.976753S.D.dependent var0.093396S.E. of regressi on0.014240Akaike info-5.459968criteri onSum squared resid0.003853Schwarz criteri on-5.167438Log likelihood74.24960F-statistic202.6826Durb in-Watson stat1.791427Prob(F

18、-statistic)0.000000得出回归方程为:LOG(Y) = 0.381144581612*LOG(X1) + 1.22228859801*LOG(X2) - 0.0811098881534*L0G(X3)- 0.04722870996*L0G(X4) - 0.101173736285*LOG(X5) - 4过检验结果可知 RR较大且接近于1,而且F=202.6826 > Fo.o55,19 = 2.74 ,故认为粮食产量与上述变量之间总体线性关系显著.但是由于其中X4、X5前的参数估计值未通过t检验,且符号的经济意义不合理,故认为解释变量之间存在多重

19、共线2、相关系数表LNX1LNX2LNX3LNX4LNX5LNX11.000000-0.5687440.4517000.9643570.440205LNX2-0.5687441.000000-0.214097-0.697625-0.073270LNX30.451700-0.2140971.0000000.3987800.411279LNX40.964357-0.6976250.3987801.0000000.279528LNX50.440205-0.0732700.4112790.2795281.000000由表可知LnX与LnX2之间存在高度的线性相关性3、简单的回归形式LnY 与 LnXD

20、epe ndent Variable: LNYMethod: Least SquaresDate: 12/11/14 Time: 15:15Sample: 1983 2007In cluded observati ons: 25VariableCoefficie ntStd. Error t-StatisticProb.LNX10.2240050.0255158.7792930.0000C8.9020210.20603443.206570.0000R-squared0.770175Mea n depe ndent var10.70905Adjusted R-squared0.760182S.D

21、.dependent var0.093396S.E. of regressi on0.045737Akaike info-3.255189criteri onSum squared resid0.048114Schwarz criteri on-3.157679Log likelihood42.68986F-statistic77.07599Durb in-Watson stat0.939435Prob(F-statistic)0.000000LnY 与 LnX2Depe ndent Variable: LNYMethod: Least SquaresDate: 12/11/14 Time:

22、15:16Sample: 1983 2007In cluded observati ons: 25VariableCoefficie ntStd. Error t-StatisticProb.LNX2-0.3834340.509669-0.7523210.4595C15.157485.9129712.5634290.0174R-squared0.024017Mean depe ndent var10.70905Adjusted R-squared-0.018417S.D.dependent var0.093396S.E. of regressi on0.094252Akaike info-1.

23、809063criteri onSum squared resid0.204321Schwarz criteri on-1.711553Log likelihood24.61329F-statistic0.565986Durb in-Watson stat0.335219Prob(F-statistic)0.459489LnY 与 LnX3Depe ndent Variable: LNYMethod: Least SquaresDate: 12/11/14 Time: 15:18Sample: 1983 2007In cluded observati ons: 25VariableCoeffi

24、cie ntStd. Error t-StatisticProb.LNX30.1080670.0852711.2673350.2177C9.6197220.85974411.189050.0000R-squared0.065274Mean depe ndent var10.70905Adjusted R-squared0.024634S.D.dependent var0.093396S.E. of regressi on0.092239Akaike info-1.852255criteri onSum squared resid0.195684Schwarz criteri on-1.7547

25、45Log likelihood25.15319F-statistic1.606139Durb in-Watson stat0.597749Prob(F-statistic)0.217717LnY 与 LnX4Depe ndent Variable: LNYMethod: Least SquaresDate: 12/11/14 Time: 15:18Sample: 1983 2007In cluded observati ons: 25VariableCoefficie ntStd. Error t-StatisticProb.LNX40.1669760.0282745.9056700.000

26、0C8.9490900.29825530.004790.0000R-squared0.602605Mean depe ndent var10.70905Adjusted R-squared0.585327S.D.dependent var0.093396S.E. of regressi on0.060143Akaike info-2.707578criteri onSum squared resid0.083194Schwarz criteri on-2.610068Log likelihood35.84472F-statistic34.87693Durb in-Watson stat0.62

27、5528Prob(F-statistic)0.000005LnY 与 Ln%Depe nde nt Variable: LNYMethod: Least SquaresDate: 12/11/14 Time: 15:19Sample: 1983 2007In cluded observati ons: 25VariableCoefficie ntStd. Error t-StatisticProb.LNX50.4887310.2346062.0831990.0485C5.6007492.4522072.2839620.0319R-squared0.158733Mean depe ndent v

28、ar10.70905Adjusted R-squared0.122156S.D.dependent var0.093396S.E. of regressi onSum squared residLog likelihoodDurb in-Watson stat-1.957599-1.8600894.3397180.0485380.087506Akaike infocriteri on0.176118Schwarz criterion26.46999F-statistic0.327932Prob(F-statistic)比拟各个回归方程的R2可知Y与X1的氏最大,即粮食生产受农业化肥施用量最大,

29、 与经验相符,因此选为初始的回归方程.且初始化回归方程为:LOG(Y) = 0.224004867873*LOG(X1) + 8.90202121784R = 0.770175D.W. = 0.9394354、逐步回归LnY 与 LnXDepe ndent Variable: LNYMethod: Least SquaresDate: 12/11/14 Time: 15:28Sample: 1983 2007In cluded observati ons: 25VariableCoefficie ntStd. Error t-StatisticProb.LNX10.2240050.025515

30、8.7792930.0000C8.9020210.20603443.206570.0000R-squared0.770175Mea n depe ndent var10.70905Adjusted R-squared S.E. of regressi onSum squared residLog likelihoodDurb in-Watson stat0.760182 S.D. depe nde nt var0.045737 Akaike infocriteri on0.048114 Schwarz criterio n0.093396-3.255189-3.15767942.689860.

31、939435F-statisticProb(F-statistic)77.075990.000000LnY 与 LnX、LnXDepe ndent Variable: LNYMethod: Least SquaresDate: 12/11/14 Time: 15:29Sample: 1983 2007In cluded observati ons: 25VariableCoefficie ntStd. Error t-StatisticProb.LNX10.2978540.01548219.239290.0000LNX21.2586220.1500668.3871270.0000C-6.295

32、6821.814941-3.4688090.0022R-squared0.945246Mean depe ndent var10.70905Adjusted R-squared0.940269S.D.dependent var0.093396S.E. of regressi on0.022826Akaike info-4.609666criteri onSum squared resid0.011463Schwarz criteri on-4.463401Log likelihood60.62083F-statistic189.9002Durb in-Watson stat1.595748Pr

33、ob(F-statistic)0.000000由输出结果可知 氏有所提升,且各解释变量前得参数均通过 t检验,符号也合理D.W枪验也说明不存在一阶自相关.可以考虑再此模型上继续引入X3LnY 与 LnX、Ln关、LnXDepe ndent Variable: LNYMethod: Least SquaresDate: 12/11/14 Time: 15:30Sample: 1983 2007In eluded observati ons: 25VariableCoefficie ntStd. Error t-StatisticProb.LNX10.3233850.01086129.775520

34、.0000LNX21.2907290.09615313.423650.0000LNX3-0.0867540.015155-5.7244840.0000C-5.9996381.162078-5.1628520.0000R-squared0.978616Mean depe ndent var10.70905Adjusted R-squared0.975561S.D.dependent var0.093396S.E. of regressi on0.014601Akaike info-5.469854criteri onSum squared resid0.004477Schwarz criteri

35、 on-5.274834Log likelihood72.37318F-statistic320.3438Durb in-Watson stat1.412883Prob(F-statistic)0.000000由输出结果可知R2再次提升且参数符号合理,变量通过t检验.但是D.W.=1.419(d>_=1.12、du=1.66)落入无法判断的区域,且 X4的参数没有通过t检验LM检验Breusch-Godfrey Serial Correlation LM Test:F-statistic1.241319Probability0.278428Obs*R-squared1.460972Pro

36、bability0.226776Test Equati on:Depe ndent Variable: RESIDMethod: Least SquaresDate: 12/11/14 Time: 15:43VariableCoefficie ntStd. Error t-StatisticProb.LNX10.0024030.0110120.2182250.8295LNX20.0069520.0958090.0725610.9429LNX3-0.0054780.015850-0.3455890.7333C-0.0447291.156156-0.0386880.9695RESID(-1)0.2

37、574590.2310821.1141450.2784R-squared0.058439Mean depe ndent var1.07E-16Adjusted R-squared-0.129873S.D.dependent var0.013658S.E. of regressi on0.014517Akaike info-5.450070criteri onSum squared resid0.004215Schwarz criteri on-5.206295Log likelihood73.12588F-statistic0.310330Durb in-Watson stat1.794969

38、Prob(F-statistic)0.867655LM检验显示不存在一阶自相关,继续引入 XtLnY 与 LnX、LnX2、Ln%、LnXtDepe ndent Variable: LNYMethod: Least SquaresDate: 12/11/14 Time: 15:32Sample: 1983 2007In cluded observati ons: 25VariableCoefficien Std. Error t-StatisticProb.tLNX10.3220610.0391618.2239570.0000LNX21.2940010.1353689.5591170.0000

39、LNX3-0.0866650.015730-5.5095090.0000LNX40.0013030.0369720.0352510.9722C-6.0415541.682783-3.5902150.0018R-squared0.978617Mean depe ndent var10.70905Adjusted R-squared0.974341S.D.dependent var0.093396S.E. of regressi on0.014961Akaike info-5.389916criteri onSum squared resid0.004476Schwarz criteri on-5.146141Log likelihood72.37395F-statistic228.8316Durb in-Watson stat1.413284Prob(F-statistic)0.000000由输出结果可知R2有所下降,且X4的参数未能通过t检验.去掉X4引入X5LnY 与 LnX1、LnX2、LnX、LnXsDepe ndent Variable: LNYMethod: Least SquaresDate: 12/11/14 Time: 15:33Sample: 1983 2007In cluded o