收藏 分销(赏)

计量经济学实验报告csust.doc

上传人:人****来 文档编号:3768736 上传时间:2024-07-17 格式:DOC 页数:27 大小:806.04KB
下载 相关 举报
计量经济学实验报告csust.doc_第1页
第1页 / 共27页
计量经济学实验报告csust.doc_第2页
第2页 / 共27页
计量经济学实验报告csust.doc_第3页
第3页 / 共27页
计量经济学实验报告csust.doc_第4页
第4页 / 共27页
计量经济学实验报告csust.doc_第5页
第5页 / 共27页
点击查看更多>>
资源描述

1、 实验报告课程名称: 计量经济学 实验项目: 计量经济学Eviews应用与操作 学生姓名: 学 号: 班 级: 专 业: 指导教师: 2023年12月指导老师评语:签字:年月日成绩等级:备注:实验任务一1. 根据数据1构建截面数据一元线性模型。假设拟建立如下一元回归模型: Y= 下图为用Eviews 软件对数据进行回归分析的计算结果:Dependent Variable: YMethod: Least SquaresDate: 01/07/03 Time: 23:44Sample: 1 31Included observations: 31CoefficientStd. Errort-Stat

2、isticProb.X1.3594770.04330231.395250.0000C-57.90655377.7595-0.1532890.8792R-squared0.971419Mean dependent var11363.69Adjusted R-squared0.970433S.D. dependent var3294.469S.E. of regression566.4812Akaike info criterion15.57911Sum squared resid9306127.Schwarz criterion15.67162Log likelihood-239.4761Han

3、nan-Quinn criter.15.60926F-statistic985.6616Durbin-Watson stat1.294974Prob(F-statistic)0.000000由此可得: (377.76) (0.043) (-0.15) (31.40) R2=0.972. 对模型进行检查。从回归估计的结果看,模型拟合较好。可决系数R2=0.97,拟合优度较高,表白该地区消费支出变化的97%可以由该地区可支配收入的变化来解释。从斜率项的t检查值来看,在1%的水平上通过了显著性检查,它表白,人均可支配收入每增长1元,人均消费支出增长1.36元。但是斜率值1.361,不符合经济规律。3

4、. 若2023年某地区人均可支配收入为4100元,那么该地区消费支出是多少? (元)实验任务二1. 根据数据2构建时间数据一元线性模型。假设拟建立如下一元回归模型: Y=下图为用Eviews 软件对数据进行回归分析的计算结果:Dependent Variable: YMethod: Least SquaresDate: 01/31/02 Time: 15:25Sample: 1978 2023Included observations: 29CoefficientStd. Errort-StatisticProb.X0.4375270.00929747.059500.0000C2091.295

5、334.98696.2429140.0000R-squared0.987955Mean dependent var14855.72Adjusted R-squared0.987509S.D. dependent var9472.076S.E. of regression1058.633Akaike info criterion16.83382Sum squared resid30259014Schwarz criterion16.92811Log likelihood-242.0903Hannan-Quinn criter.16.86335F-statistic2214.596Durbin-W

6、atson stat0.277155Prob(F-statistic)0.000000由此可得: (334.99) (0.009) (6.24) (47.06)R2=0.992. 对模型进行检查。 从回归估计的结果看,可决系数R2=0.99,拟合优度较高,模型拟合较好,表白实际消费支出的变化的99%可以由实际可支配收入的变化来解释。从斜率项的t检查值来看,在1%的水平上通过了显著性检查,且斜率项00.441,符合经济规律,表白人均可支配收入每增长1亿元,消费支出增长0.44亿元。 3. 若2023年我国可支配总收入为54180亿元,那么该相应的总消费是多少? (亿元)实验任务三1. 根据数据3

7、构建截面数据多元线性模型。 假设拟建立如下多元截面数据模型: Y=下图为用Eviews 软件对数据进行回归分析的计算结果:Dependent Variable: YMethod: Least SquaresDate: 06/09/15 Time: 22:08Sample: 1 31Included observations: 31CoefficientStd. Errort-StatisticProb.X10.5556440.0753087.3783200.0000X20.2500850.1136342.2023910.0362C143.3265260.40320.5504020.5864R-

8、squared0.975634Mean dependent var8401.468Adjusted R-squared0.973893S.D. dependent var2388.459S.E. of regression385.9169Akaike info criterion14.84089Sum squared resid4170093.Schwarz criterion14.97966Log likelihood-227.0337Hannan-Quinn criter.14.88612F-statistic560.5650Durbin-Watson stat1.843488Prob(F

9、-statistic)0.000000散点图:表白2023年可支配收入X1与Y存在线性正相关关系,并且,2023年消费支出X2与Y存在线性正相关关系,这表白居民消费支出不仅受本年可支配收入的影响,也受上一年消费支出的影响,即存在棘轮效应。 估计方程: (0.55) (7.38)(2.20)2. 对模型进行检查。从回归估计结果看出,R2=0.98,这说明拟合优度高,模型拟合较好,表白2023年消费支出变化的98%可以由2023年可支配收入X1和2023年消费支出X2来解释。从回归模型的t检查值来看,X1在1%的水平上通过了显著性检查,X2在5%的水平上通过了显著性检查,可判断X1和X2对Y均有显

10、著影响。从回归模型的F检查值来看,F=560.57,其随着概率为零,在1%的水平上通过显著性检查,说明回归方程显著。斜率项00.561,00.251,符合经济规律。这表白了在2023年可支配收入X1保持不变的情况下,每增长1元2023年消费支出X2,2023年消费支出Y变动0.25元;在2023年消费支出X2保持不变的情况下,每增长1元2023年可支配收入X1,2023年消费支出Y变动0.56元。实验任务四1. 根据数据4构建时间序列数据多元线性模型。根据需求理论,P0为食品价格,P1为通货膨胀率,X为食品消费支出总额。Q=f(X, P0 ,P1), 用Eviews软件对数据进行回归分析,结果

11、如下: Dependent Variable: QMethod: Least SquaresDate: 06/10/15 Time: 02:02Sample: 1985 2023Included observations: 22CoefficientStd. Errort-StatisticProb.X0.2102060.01175117.888570.0000P06.6803343.3066302.0202850.0585P1-5.8547232.929604-1.9984690.0610C877.204137.0912423.649900.0000R-squared0.982629Mean

12、 dependent var1830.000Adjusted R-squared0.979734S.D. dependent var365.1392S.E. of regression51.98063Akaike info criterion10.90258Sum squared resid48635.74Schwarz criterion11.10096Log likelihood-115.9284Hannan-Quinn criter.10.94932F-statistic339.4076Durbin-Watson stat0.737832Prob(F-statistic)0.000000

13、 2. 对模型进行检查。 可决系数R2=0.9826,,拟合优度较高,模型拟合好。 从t检查值看,解释变量X、P0、P1分别在1%、10%、10%的水平上通过了显著性检查。 F值=339.41,所相应的随着概率为0,小于1%,表白整体模型在1%的水平上通过了显著性检查。 P1的斜率项为-5.85,与理论相悖,是由于随着通货膨胀率的增长,实际收入水平会下降,食品消费支出减少,食品需求减少。P0斜率项为正,说明随着食品价格的增长,消费支出也会增长,此商品为吉芬商品,符合经济规律。 实验任务五1. 估计中国农村居民人均消费函数。 假设拟建立如下回归模型: =+X1+X2+下图为用Eviews 软件对

14、数据进行回归分析的计算结果:Dependent Variable: YMethod: Least SquaresDate: 05/25/15 Time: 03:06Sample: 1 31Included observations: 31CoefficientStd. Errort-StatisticProb.X10.4020970.1648942.4385140.0213X20.7090300.04171016.999110.0000C728.1402328.15582.2188860.0348R-squared0.922173Mean dependent var2981.623Adjust

15、ed R-squared0.916614S.D. dependent var1368.763S.E. of regression395.2538Akaike info criterion14.88870Sum squared resid4374316.Schwarz criterion15.02747Log likelihood-227.7748Hannan-Quinn criter.14.93394F-statistic165.8853Durbin-Watson stat1.428986Prob(F-statistic)0.000000用普通最小二乘法的估计结果如下: =728.14+0.4

16、0 X1+0.71X2 (2.22) (2.44) (17.00) =0.92 D.W.=1.43 F=165.89 1 对模型进行异方差检查。2.1图示法不同地区农村人均消费支出的差别重要来源于非农经营收入及工资收入、财政收入等其他收入的差别上,因此,假如存在异方差性,则也许是X2引起的。上图我们得到了残差平方项与X2的散点图,残差平方项在不断变化,存在明显的散点扩大趋势,且随着X2的增大,残差平方项也在不断增大,因此存在递增型异方差性。 2.2 G-Q检查Heteroskedasticity Test: Breusch-Pagan-GodfreyF-statistic8.048853Pro

17、b. F(2,28)0.0017Obs*R-squared11.31644Prob. Chi-Square(2)0.0035Scaled explained SS23.78437Prob. Chi-Square(2)0.0000Test Equation:Dependent Variable: RESID2Method: Least SquaresDate: 06/10/15 Time: 13:57Sample: 1 31Included observations: 31CoefficientStd. Errort-StatisticProb.C-246960.3222964.1-1.1076

18、230.2774X188.52762112.03690.7901650.4361X2108.509628.339553.8289090.0007R-squared0.365046Mean dependent var141107.0Adjusted R-squared0.319692S.D. dependent var325595.5S.E. of regression268553.6Akaike info criterion27.93125Sum squared resid2.02E+12Schwarz criterion28.07003Log likelihood-429.9344Hanna

19、n-Quinn criter.27.97649F-statistic8.048853Durbin-Watson stat2.178345Prob(F-statistic)0.001731 由此可得,在1%的显著性水平下拒绝同方差的原假设,该模型存在异方差。2.3怀特检查Heteroskedasticity Test: WhiteF-statistic3.898573Prob. F(5,25)0.0095Obs*R-squared13.58148Prob. Chi-Square(5)0.0185Scaled explained SS28.54493Prob. Chi-Square(5)0.000

20、0Test Equation:Dependent Variable: RESID2Method: Least SquaresDate: 05/25/15 Time: 03:11Sample: 1 31Included observations: 31CoefficientStd. Errort-StatisticProb.C1062731.993615.41.0695590.2950X1-902.44931056.763-0.8539750.4012X120.1852650.2692490.6880790.4977X1*X20.2467970.1389101.7766650.0878X2-52

21、2.6261363.4949-1.4377810.1629X220.0455460.0293261.5531150.1330R-squared0.438112Mean dependent var141107.0Adjusted R-squared0.325735S.D. dependent var325595.5S.E. of regression267358.4Akaike info criterion28.00255Sum squared resid1.79E+12Schwarz criterion28.28010Log likelihood-428.0396Hannan-Quinn cr

22、iter.28.09303F-statistic3.898573Durbin-Watson stat2.151382Prob(F-statistic)0.009480 由此可得,模型在1%的显著性水平下拒绝同方差的原假设,该模型存在异方差。3. 采用加权最小二乘法估计方程Dependent Variable: YMethod: Least SquaresDate: 05/25/15 Time: 03:27Sample: 1 31Included observations: 31Weighting series: 1/ABS(RESID)CoefficientStd. Errort-Statis

23、ticProb.X10.4710240.03557013.242210.0000X20.6912980.01162859.451730.0000C642.164056.7915711.307380.0000Weighted StatisticsR-squared0.992927Mean dependent var2595.506Adjusted R-squared0.992422S.D. dependent var3551.699S.E. of regression67.05808Akaike info criterion11.34076Sum squared resid125910.0Sch

24、warz criterion11.47953Log likelihood-172.7818Hannan-Quinn criter.11.38600F-statistic1965.400Durbin-Watson stat1.604499Prob(F-statistic)0.000000Unweighted StatisticsR-squared0.920239Mean dependent var2981.623Adjusted R-squared0.914424S.D. dependent var1368.763S.E. of regression400.4084Sum squared res

25、id4489153.Durbin-Watson stat1.580905 由此可得: (11.31)(13.24)(59.45)R2=0.99通过比较,可以发现在加权后,R2值增长了,模型的拟合优度提高,消除了异方差。同时X1,X2的t记录量有了显著改善,都在1%的限度上通过了显著性检查,更好拟合了原模型。实验任务六1. 对实验数据3各变量取对数,并估计方程 ,判断其是否存在序列相关性。下图为用Eviews 软件对数据进行回归分析的计算结果:Dependent Variable: LNYMethod: Least SquaresDate: 06/10/15 Time: 15:28Sample:

26、 1980 2023Included observations: 28CoefficientStd. Errort-StatisticProb.LNX0.8544150.01421960.090580.0000C1.5884780.13422011.834920.0000R-squared0.992851Mean dependent var9.552256Adjusted R-squared0.992576S.D. dependent var1.303948S.E. of regression0.112351Akaike info criterion-1.465625Sum squared r

27、esid0.328192Schwarz criterion-1.370468Log likelihood22.51875Hannan-Quinn criter.-1.436535F-statistic3610.878Durbin-Watson stat0.379323Prob(F-statistic)0.000000由此可得:(11.83)(60.09)R2=0.99 D.W=0.3793 F=3610.882.对模型进行序列相关性检查。2.1图示法 由图:这期的残差e随着上一期的残差e1的增长而增长,存在正序列相关性。2.2D.W.检查法 由于D.W=0.3793, n=28, k=2,查D

28、.W分布表可得,dL=1.33, du=1.48,0D.W1.33,所以存在一阶正自相关。2.3拉格朗日乘数(LM)检查法(1)1阶自相关检查:如下图,只看rob. Chi-Square(1)=0.0001,因此在1%的水平上拒绝不存在自相关的原假设,表白存在一阶自相关。Breusch-Godfrey Serial Correlation LM Test:F-statistic32.78471Prob. F(1,25)0.0000Obs*R-squared15.88607Prob. Chi-Square(1)0.0001Test Equation:Dependent Variable: RES

29、IDMethod: Least SquaresDate: 06/10/15 Time: 16:28Sample: 1980 2023Included observations: 28Presample missing value lagged residuals set to zero.CoefficientStd. Errort-StatisticProb.INX-0.0028360.009551-0.2969270.7690C0.0233450.0901240.2590330.7977RESID(-1)0.7697160.1344305.7257930.0000R-squared0.567

30、360Mean dependent var7.33E-16Adjusted R-squared0.532748S.D. dependent var0.110251S.E. of regression0.075363Akaike info criterion-2.232045Sum squared resid0.141989Schwarz criterion-2.089309Log likelihood34.24863Hannan-Quinn criter.-2.188409F-statistic16.39235Durbin-Watson stat1.042286Prob(F-statistic

31、)0.000028(2)一直检查到15阶,随着概率11.14%才大于10%,没有通过检查,这表白模型存在直到15阶的自相关。下图为15阶自相关检查。Breusch-Godfrey Serial Correlation LM Test:F-statistic2.612572Prob. F(15,11)0.0568Obs*R-squared21.86315Prob. Chi-Square(15)0.1114Test Equation:Dependent Variable: RESIDMethod: Least SquaresDate: 06/10/15 Time: 16:48Sample: 198

32、0 2023Included observations: 28Presample missing value lagged residuals set to zero.CoefficientStd. Errort-StatisticProb.INX-0.0021080.011190-0.1883620.8540C0.0135300.1037070.1304660.8986RESID(-1)1.0446830.2987283.4971020.0050RESID(-2)-0.6892050.434177-1.5873820.1407RESID(-3)0.3008100.4825390.623389

33、0.5457RESID(-4)-0.4679020.484487-0.9657670.3549RESID(-5)0.4257730.5048990.8432830.4170RESID(-6)-0.3247090.519778-0.6247060.5449RESID(-7)-0.0283210.530359-0.0534000.9584RESID(-8)-0.1106780.531208-0.2083520.8388RESID(-9)0.0829660.5409900.1533600.8809RESID(-10)-0.2237300.553029-0.4045530.6936RESID(-11)

34、0.0322670.5699230.0566170.9559RESID(-12)-0.3127620.549093-0.5695970.5804RESID(-13)-0.0303290.555608-0.0545860.9574RESID(-14)0.0273480.4941170.0553470.9569RESID(-15)-0.1786250.354512-0.5038630.6243R-squared0.780827Mean dependent var7.33E-16Adjusted R-squared0.462029S.D. dependent var0.110251S.E. of r

35、egression0.080865Akaike info criterion-1.912089Sum squared resid0.071931Schwarz criterion-1.103251Log likelihood43.76925Hannan-Quinn criter.-1.664819F-statistic2.449286Durbin-Watson stat1.536255Prob(F-statistic)0.0683063.采用广义差分法估计方程(1)一阶广义差分:如下图,D.W=0.83,n=27,k=3,查表,得到临界值dL=1.02 和du=1.32, 0D.W1.02,存

36、在正自相关。Dependent Variable: INYMethod: Least SquaresDate: 06/10/15 Time: 17:07Sample (adjusted): 1981 2023Included observations: 27 after adjustmentsConvergence achieved after 200 iterationsCoefficientStd. Errort-StatisticProb.INX0.4963880.1005394.9372500.0000C333.655840410.850.0082570.9935AR(1)0.9998320.02077048.138720.0000R-squared0.998115Mean depen

展开阅读全文
部分上传会员的收益排行 01、路***(¥15400+),02、曲****(¥15300+),
03、wei****016(¥13200+),04、大***流(¥12600+),
05、Fis****915(¥4200+),06、h****i(¥4100+),
07、Q**(¥3400+),08、自******点(¥2400+),
09、h*****x(¥1400+),10、c****e(¥1100+),
11、be*****ha(¥800+),12、13********8(¥800+)。
相似文档                                   自信AI助手自信AI助手
百度文库年卡

猜你喜欢                                   自信AI导航自信AI导航
搜索标签

当前位置:首页 > 包罗万象 > 大杂烩

移动网页_全站_页脚广告1

关于我们      便捷服务       自信AI       AI导航        获赠5币

©2010-2024 宁波自信网络信息技术有限公司  版权所有

客服电话:4008-655-100  投诉/维权电话:4009-655-100

gongan.png浙公网安备33021202000488号   

icp.png浙ICP备2021020529号-1  |  浙B2-20240490  

关注我们 :gzh.png    weibo.png    LOFTER.png 

客服