2022年《计量经济学综合实验》实验报告

计量经济学综合实验实验报告2013-2014学年第一学期班级：姓名：学号：课程编码：0123100320课程类型：综合实训实验时间：第16周至第18周实验地点：实验目的和要求：熟悉eviews软件的根本功能，能运用eviews软件进展一元和多元模型的参数可能、统计检验和预测分析，能运用eviews软件进展异方差、自相关、多重共线性的检验和处理，并最终将操作结果进展分析。能熟悉运用eviews软件对时间序列进展单位根、协整和格兰杰因果关系检验。实验所用软件：eviews 实验内容和结论：见第2页第39页计量经济学综合实验实验一第二章第6题Dependent Variable: YMethod: Least SquaresDate: 12/17/13 Time: 09:13Sample: 1985 1998Included observations: 14VariableCoefficientStd. Errort-StatisticProb. C12596.271244.56710.121010.0000GDP26.954154.1203006.5417920.0000R-squared0.781002 Mean dependent var20168.57Adjusted R-squared0.762752 S.D. dependent var3512.487S.E. of regression1710.865 Akaike info criterion17.85895Sum squared resid35124719 Schwarz criterion17.95024Log likelihood-123.0126 F-statistic42.79505Durbin-Watson stat0.859998 Prob(F-statistic)0.000028（1） （10.12） （6.54） （2）是样本回归方程的斜率，它表示GDP每增加1亿元，物资运输量将增加26.95万吨，是样本回归方程的截距，表示GDP不变价时的物资运输量。（3），说明离差平方和的78%被样本回归直线解释，还有22%未被解释。因而，样本回归至西安对样本点的拟合优度是较高的。给出明显水平，查自由度v=14-2=12的t分布表，得临界值，故回归系数均明显不为零，回归模型中英包含常数项，X对Y有明显妨碍。（4）2000年的国内消费总值为620亿元，物资运输量预测值为29307.84万吨。 实验二第二章第7题X1Dependent Variable: QMethod: Least SquaresDate: 12/17/13 Time: 10:57Sample: 1978 1998Included observations: 21VariableCoefficientStd. Errort-StatisticProb. C40772.471389.79529.337040.0000X10.0012200.0019090.6391940.5303R-squared0.021051 Mean dependent var40996.12Adjusted R-squared-0.030473 S.D. dependent var6071.868S.E. of regression6163.687 Akaike info criterion20.38113Sum squared resid7.22E+08 Schwarz criterion20.48061Log likelihood-212.0019 F-statistic0.408568Durbin-Watson stat0.206201 Prob(F-statistic)0.530328=40772.47+0.001+X2Dependent Variable: QMethod: Least SquaresDate: 12/17/13 Time: 10:58Sample: 1978 1998Included observations: 21VariableCoefficientStd. Errort-StatisticProb. C26925.65915.865729.399120.0000X25.9125340.35642316.588510.0000R-squared0.935413 Mean dependent var40996.12Adjusted R-squared0.932014 S.D. dependent var6071.868S.E. of regression1583.185 Akaike info criterion17.66266Sum squared resid47623035 Schwarz criterion17.76214Log likelihood-183.4579 F-statistic275.1787Durbin-Watson stat1.264400 Prob(F-statistic)0.000000=26925.65+5.91+X3Dependent Variable: QMethod: Least SquaresDate: 12/17/13 Time: 10:58Sample: 1978 1998Included observations: 21VariableCoefficientStd. Errort-StatisticProb. C-49865.3912638.40-3.9455450.0009X31.9487000.2706347.2004980.0000R-squared0.731817 Mean dependent var40996.12Adjusted R-squared0.717702 S.D. dependent var6071.868S.E. of regression3226.087 Akaike info criterion19.08632Sum squared resid1.98E+08 Schwarz criterion19.18580Log likelihood-198.4064 F-statistic51.84718Durbin-Watson stat0.304603 Prob(F-statistic)0.000001=-49865.39+1.95+（1） =40772.47+0.001+ =26925.65+5.91+ =-49865.39+1.95+（2）=0.001为样本回归方程的斜率，表示边际农业机械总动力，说明农业机械总动力每增加1万千瓦，粮食产量增加1万吨。=40072.47是截距，表示不受农业机械总动力妨碍的粮食产量。=0.02，说明总离差平方和的2%被样本回归直线解释，有98%未被解释，因而样本回归直线对样本点的拟合优度是非常低的。给出的明显水平=0.05，查自由度v=21-2=19的t分布表，得临界值， ，=16.6,故回归系数均不为零，回归模型中应包含常数项，X对Y有明显妨碍。=1.95为样本回归方程的斜率，表示边际土地灌溉面积，说明土地灌溉面积每增加1千公顷，粮食产量增加1万吨。=-49865.39是截距，表示不受土地灌溉面积妨碍的粮食产量。=0.73，说明总离差平方和的73%被样本回归直线解释，有27%未被解释，因而样本回归直线对样本点的拟合优度是较高的。给出明显性水平=0.05，查自由度=21-2=19的t分布表，得临界值=2.09，=-3.95,故回归系数包含零，回归模型中不应包含常数项，X对Y有无明显妨碍。（3）依照分析，X2得拟合优度最高，模型最好，因而选择X2得预测值。=26925.65+5.91+实验三P85第3题Dependent Variable: YMethod: Least SquaresDate: 12/19/13 Time: 09:10Sample: 1 18Included observations: 18VariableCoefficientStd. Errort-StatisticProb. C-0.97556830.32236-0.0321730.9748X1104.31466.40913616.275920.0000X20.4021900.1163483.4567760.0035R-squared0.979727 Mean dependent var755.1500Adjusted R-squared0.977023 S.D. dependent var258.6859S.E. of regression39.21162 Akaike info criterion10.32684Sum squared resid23063.27 Schwarz criterion10.47523Log likelihood-89.94152 F-statistic362.4430Durbin-Watson stat2.561395 Prob(F-statistic)0.000000（1）（2）提出检验的原假设为。给出明显水平，查自由度v=18-2=16的t分布表，得临界值。，因而否认,明显不等于零，即能够认为受教育年限对购置书籍及课外读物支出有明显妨碍。，因而否认，明显不等于零，即能够家庭月可支配收入对购置书籍及课外读物支出有明显妨碍。（3） =0.9797，表示Y中的变异功能被可能的回归方程解释的部分越多，可能的回归方程对样本观测值就拟合的越好。同样，=0.9770，非常接近1，表示模型拟合度非常好。（4）把=10，=480代入实验四P86第6题Dependent Variable: YMethod: Least SquaresDate: 12/19/13 Time: 10:14Sample: 1955 1984Included observations: 30VariableCoefficientStd. Errort-StatisticProb. C0.2089324.3722180.0477860.9623X11.0814070.2341394.6186490.0001X23.6465651.6998492.1452290.0414X30.0042120.0116640.3610710.7210R-squared0.552290 Mean dependent var22.13467Adjusted R-squared0.500632 S.D. dependent var14.47115S.E. of regression10.22618 Akaike info criterion7.611345Sum squared resid2718.944 Schwarz criterion7.798171Log likelihood-110.1702 F-statistic10.69112Durbin-Watson stat1.250501 Prob(F-statistic)0.000093，表示该地区某农产品收买量随着销售量的增加而增加，=3.647表示农产品收买量随出口量的增加而增加。=3.647表示农产品收买量随库存量的增加而增加。该回归方程系数的符号和大小均符合经济理论和实际情况。统计检验a回归方程的明显性检验F检验：r=0.55表示和和结合起来对Y的解释才能到达55，因而，样本回归方程的拟合优度是高的。明显性水平=0.05，查自由度v=30-3-1=27，的F分布表的临界值(3，27)=2.96，F=10.69F(3，27)=2.96,说明回归方程在总体上是明显的。b回归系数的明显性检验 t检验：明显性水平=0.05，查自由度v=30-3-1=26的t分布表的临界值t(26)=2.06，t=4.62t(26)，因而明显不为零，即销售量对农产品收买量有明显妨碍；t=2.15t(26)，因而明显不为零，即出口量对农产品收买量有明显妨碍；t=0.36，=81.21故明显不为零，则回归模型中应包含常数项，能够认为时间对销售额有明显妨碍，,表示Y能对可能的回归方程进展非常高解释，因而可能的回归方程对样本观测值就拟合的程度非常高T=10，Lny=3.949y=49.4024则预测得该商场1999年的皮鞋销售额为49.4024万元实验六P107第四章第2题Dependent Variable: LOGYMethod: Least SquaresDate: 12/20/13 Time: 15:08Sample: 1 21Included observations: 21VariableCoefficientStd. Errort-StatisticProb. C-35.404251.637922-21.615350.0000T0.0207660.00086623.974010.0000R-squared0.968000 Mean dependent var3.843167Adjusted R-squared0.966316 S.D. dependent var1.309610S.E. of regression0.240355 Akaike info criterion0.076997Sum squared resid1.097644 Schwarz criterion0.176475Log likelihood1.191533 F-statistic574.7531Durbin-Watson stat0.110127 Prob(F-statistic)0.000000LnY=-35.4042+0.0208+Lnyf=6.127Y=458.0599实验七P108第四章第3题Dependent Variable: LNMMethod: Least SquaresDate: 12/20/13 Time: 16:35Sample: 1948 1964Included observations: 17VariableCoefficientStd. Errort-StatisticProb. LNP1.2658790.4313932.9344020.0116LNR0.8645950.5172281.6715930.1185LNY0.2062100.3087200.6679520.5158C-2.0950901.790906-1.1698500.2631R-squared0.859355 Mean dependent var5.481567Adjusted R-squared0.826899 S.D. dependent var0.269308S.E. of regression0.112047 Akaike info criterion-1.337475Sum squared resid0.163208 Schwarz criterion-1.141425Log likelihood15.36854 F-statistic26.47717Durbin-Watson stat0.743910 Prob(F-statistic)0.000008ln (-1.1699) (2.9344) (1.6716) (0.6680) （2）t检验：假设:，明显性水平=0.05，查自由度v=17-3-1=13的t分布表的临界值t(13)=2.16，t=2.9344t(13)，因而明显不为零，即内含价格缩减指数对名义货币存量有明显妨碍；=1.6716t(13)，因而明显为零，即长期利率对名义货币存量无明显妨碍；F(3，13)=3.41,因而否认，说明回归方程在总体上是明显的。即内含价格缩减指数，名义国名收入和长期利率与名义货币存量之间的关系是线性的。经济意义分析：1.2659表示内含价格缩减指数每增加1%，名义货币存量就增加1.2659%，0.2062表示名义国民收入每增加1亿，名义货币存量就增加0.2062亿，0.8646表示长期利率每增加1%，名义货币存量就增加0.8646%。（3）Dependent Variable: LNMMethod: Least SquaresDate: 12/20/13 Time: 16:41Sample: 1948 1964Included observations: 17VariableCoefficientStd. Errort-StatisticProb. LNR0.9442530.4896021.9286140.0743LNY0.2265850.3000690.7551100.4627C-1.0065270.289766-3.4735840.0037R-squared0.751490 Mean dependent var0.802225Adjusted R-squared0.715989 S.D. dependent var0.205539S.E. of regression0.109537 Akaike info criterion-1.426321Sum squared resid0.167977 Schwarz criterion-1.279283Log likelihood15.12373 F-statistic21.16793Durbin-Watson stat0.656255 Prob(F-statistic)0.000059ln （-3.4736） （1.9286） （0.7551）t检验：假设:，明显性水平=0.05，查自由度v=17-2-1=14的t分布表的临界值t(14)=2.15， =1.9286t(14)，因而明显为零，即长期利率对名义货币存量有明显妨碍；=0.7551F(3，14)=3.34,因而否认，说明回归方程在总体上是明显的。即名义国名收入和长期利率与名义货币存量之间的关系是线性的。经济意义分析：0.9443表示长期利率每增加1%，名义货币存量就增加0.9443%，0.2266表示名义国民收入每增加1亿，名义货币存量就增加0.2266%。（4）Dependent Variable: LNMMethod: Least SquaresDate: 12/20/13 Time: 16:51Sample: 1948 1964Included observations: 17VariableCoefficientStd. Errort-StatisticProb. LNR-0.2094110.232757-0.8996960.3825C-1.2876770.314926-4.0888230.0010R-squared0.051201 Mean dependent var-1.569623Adjusted R-squared-0.012053 S.D. dependent var0.127733S.E. of regression0.128501 Akaike info criterion-1.155637Sum squared resid0.247686 Schwarz criterion-1.057611Log likelihood11.82291 F-statistic0.809453Durbin-Watson stat1.474376 Prob(F-statistic)0.382499ln （-4.0888）（-0.8997）t检验：假设:，明显性水平=0.05，查自由度v=17-1-1=15的t分布表的临界值t(15)=2.13，=-0.8997t(15)，因而明显为零，即长期利率对名义货币存量无明显妨碍。F检验：假设: ：r=0.05，因而，样本回归方程的拟合优度是非常低的。明显性水平=0.05，查自由度v=17-1-1=15，的F分布表的临界值(3，15)=3.29，F=0.80951.96，因而回绝原假设,接受，即等级相关系数是明显的，说明城镇居民人均生活费模型的随机误差存在异方差。（2）图示法 Y对X的散点图 残差与X的散点图（3）Dependent Variable: YMethod: Least SquaresDate: 12/26/13 Time: 10:32Sample: 1 29Included observations: 29VariableCoefficientStd. Errort-StatisticProb. C58.3179149.049351.1889640.2448X0.7955700.01837343.301930.0000R-squared0.985805 Mean dependent var2111.931Adjusted R-squared0.985279 S.D. dependent var555.5470S.E. of regression67.40436 Akaike info criterion11.32577Sum squared resid122670.4 Schwarz criterion11.42006Log likelihood-162.2236 F-statistic1875.057Durbin-Watson stat1.893970 Prob(F-statistic)0.000000White检验White Heteroskedasticity Test:F-statistic1.368420 Probability0.272237Obs*R-squared2.761902 Probability0.251339Test Equation:Dependent Variable: RESID2Method: Least SquaresDate: 12/26/13 Time: 10:34Sample: 1 29Included observations: 29VariableCoefficientStd. Errort-StatisticProb. C-22151.2616006.57-1.3838850.1782X18.1106710.958981.6525860.1104X2-0.0028580.001756-1.6273220.1157R-squared0.095238 Mean dependent var4230.013Adjusted R-squared0.025641 S.D. dependent var5479.442S.E. of regression5408.737 Akaike info criterion20.12712Sum squared resid7.61E+08 Schwarz criterion20.26856Log likelihood-288.8432 F-statistic1.368420Durbin-Watson stat1.209956 Prob(F-statistic)0.272237 （-1.3839） （1.6526） （-1.6273） T=29因而该回归模型不存在异方差。（4）戈德菲尔德-夸特检验第一个样本输出Dependent Variable: YMethod: Least SquaresDate: 12/26/13 Time: 10:49Sample: 1 11Included observations: 11VariableCoefficientStd. Errort-StatisticProb. C-287.1872271.8586-1.0563840.3183X0.9747510.1339267.2782960.0000R-squared0.854777 Mean dependent var1688.545Adjusted R-squared0.838641 S.D. dependent var122.2083S.E. of regression49.09050 Akaike info criterion10.78817Sum squared resid21688.89 Schwarz criterion10.86052Log likelihood-57.33496 F-statistic52.97359Durbin-Watson stat2.306656 Prob(F-statistic)0.000047残差平方和=21688.89第二个样本输出Dependent Variable: YMethod: Least SquaresDate: 12/26/13 Time: 10:50Sample: 19 29Included observations: 11VariableCoefficientStd. Errort-StatisticProb. C-27.68345106.7596-0.2593060.8012X0.8203370.03216925.500950.0000R-squared0.986349 Mean dependent var2641.545Adjusted R-squared0.984832 S.D. dependent var565.8140S.E. of regression69.68393 Akaike info criterion11.48878Sum squared resid43702.65 Schwarz criterion11.56113Log likelihood-61.18830 F-statistic650.2986Durbin-Watson stat2.610584 Prob(F-statistic)0.000000残差平方和=43702.65提出原假设，：备择假设，：互不一样。构造F统计量给出明显性水平=0.05，查F分布表=9，由于F=2.0153.18，因而接受原假设，即城镇居民人均生活费计量模型的随机误差不存在异方差。实验九P158第六章第3题Dependent Variable: YMethod: Least SquaresDate: 12/26/13 Time: 11:43Sample: 1975 1994Included observations: 20VariableCoefficientStd. Errort-StatisticProb. C-1.4547500.214146-6.7932610.0000X0.1762830.001445122.01700.0000R-squared0.998792 Mean dependent var24.56900Adjusted R-squared0.998725 S.D. dependent var2.410396S.E. of regression0.086056 Akaike info criterion-1.972991Sum squared resid0.133302 Schwarz criterion-1.873418Log likelihood21.72991 F-statistic14888.14Durbin-Watson stat0.734726 Prob(F-statistic)0.000000（1）线性回归模型 （-6.7933）（122.0170） s.e=0.086 DW=0.7347 T=20因而回归方程拟合效果较好，但是DW值比拟低。（2）残差图LM检验Breusch-Godfrey Serial Correlation LM Test:F-statistic11.32914 Probability0.003669Obs*R-squared7.998223 Probability0.004682Test Equation:Dependent Variable: RESIDMethod: Least SquaresDate: 12/06/13 Time: 16:38Presample missing value lagged residuals set to zero.VariableCoefficientStd. Errort-StatisticProb. C0.0609230.1716550.3549170.7270X-0.0004200.001158-0.3624390.7215RESID(-1)0.6388310.1897963.3658790.0037R-squared0.399911 Mean dependent var-8.51E-16Adjusted R-squared0.329312 S.D. dependent var0.083761S.E. of regression0.068597 Akaike info criterion-2.383669Sum squared resid0.079993 Schwarz criterion-2.234309Log likelihood26.83669 F-statistic5.664570Durbin-Watson stat1.738830 Prob(F-statistic)0.013027Obs*R-squared7.998223LM（BG）自相关检验辅助回归式可能结果是=20*0.399911=7.998223DW=1.7388由于=6.635=6.635，因而回绝，接受，因而该误差项存在一阶自相关。（4）已经知道DW=0.7347，假设给定，查表得DW检验临界值，。由于DW=0.73471.18， 依照判别规则，认为误差项不存在自相关。残差图如下：残差图经济含义是该公司的年销售额占该行业的年销售额的17.4%。实验十P159第六章第4题Dependent Variable: YMethod: Least SquaresDate: 12/09/13 Time: 08:41Sample: 1960 2001Included observations: 42VariableCoefficientStd. Errort-StatisticProb. C-3028.563655.4268-4.6207490.0000GDP0.6974920.01906036.594670.0000R-squared0.970997 Mean dependent var10765.23Adjusted R-squared0.970272 S.D. dependent var20154.12S.E. of regression3474.938 Akaike info criterion19.19099Sum squared resid4.83E+08 Schwarz criterion19.27373Log likelihood-401.0108 F-statistic1339.170Durbin-Watson stat0.178439 Prob(F-statistic)0.000000（1）线性回归模型 （-4.6207）（36.5947） s.e=3474.94 DW=0.1784 T=42因而回归方程拟合效果较好，但是DW值比拟低。（2）LM检验：Breusch-Godfrey Serial Correlation LM Test:F-statistic327.3780 Probability0.000000Obs*R-squared37.52921 Probability0.000000Test Equation:Dependent Variable: RESIDMethod: Least SquaresDate: 12/09/13 Time: 08:51Presample missing value lagged residuals set