计量经济学第八章答案第二版庞皓科学出版社.docx

1、计量经济学第八章答案第二版庞皓科学出版社我为人人，人人为我！免费分享计量经济学第二版庞皓科学出版社第八章答案 8.1 Sen和Srivastava（1971）在研究贫富国之间期望寿命的差异时，利用101个国家的数据，建立了如下的回归模型：(4.37) (0.857) (2.42) R2=0.752其中：X是以美元计的人均收入；Y是以年计的期望寿命；Sen和Srivastava 认为人均收入的临界值为1097美元（），若人均收入超过1097美元，则被认定为富国；若人均收入低于1097美元，被认定为贫穷国。括号内的数值为对应参数估计值的t-值。1）解释这些计算结果。2）回归方程中引入的原因是什么？

2、如何解释这个回归解释变量？3）如何对贫穷国进行回归？又如何对富国进行回归？4）从这个回归结果中可得到的一般结论是什么？练习题8.1参考解答：1. 结果解释 依据给定的估计检验结果数据，对数人均收入对期望寿命在统计上并没有显著影响，截距和变量在统计上对期望寿命有显著影响；同时，表明贫富国之间的期望寿命存在差异。2. 回归方程中引入的原因是从截距和斜率两个方面考证收入因素对期望寿命的影响。这个回归解释变量可解释为对期望寿命的影响存在截距差异和斜率差异的共同因素。3. 对穷国进行回归时，回归模型为对富国进行回归时，回归模型为4. 一般的结论为富国的期望寿命药高于穷国的期望寿命，并且随着收入的增加，在

3、平均意义上，富国的期望寿命的增加变化趋势优于穷国，贫富国之间的期望寿命的确存在显著差异。8.2个人所得税起征点调整对居民消费支出会产生重要的影响。为研究个人所得税起征点调整对城镇居民个人消费支出行为的效应，收集相关的数据如表8.4和表8.5所示。 表8.4 个人所得税起征点调整情况1987年1994年2006年2008年最低的起征点400元800元1600元2000元 表8.5 城镇居民收入与消费的有关数据城镇家庭平均每人可支配收入（元）城镇家庭平均每人全年消费性支出（元）平均每户城镇家庭就业人口数（人）城镇家庭平均每一就业者负担人数（含本人）（人）1985739.1673.22.151.81

4、1986900.97992.121.819871002.1884.42.091.7919881180.211042.031.7919891373.9121121.7819901510.161278.891.981.7719911700.61453.81.961.7519922026.61671.71.951.7319932577.42110.81.921.7219943496.22851.31.881.7419954282.953537.571.871.7319964838.93919.51.861.7219975160.34185.61.831.7419985425.14331.61.81.

5、75199958544615.91.771.7720006279.9849981.681.8620016859.65309.011.651.8820027702.86029.921.581.9220038472.26510.941.581.9120049421.67182.11.561.912005104937942.881.511.96200611759.458696.551.531.93200713785.819997.471.541.89200815780.811242.91.481.97若模型设定为：Consumet=Ct+1Incomet+2Consumet-1+3Employmen

6、tt+4Burdent+5d1t+6d2t+7d3t+8d4t+t其中Consumet表示t期城镇居民家庭人均消费支出，Incomet表示t期城镇居民家庭人均可支配收入,Employmentt表示t期城镇居民家庭平均每户就业人口, Burdent表示t期城镇居民家庭平均每一就业者负担人数，dit(i=1,2,3,4)相应的虚拟变量。1）构造用于描述个人所得税调整的虚拟变量，并简要说明其理由；2）用散点图描述两两变量之间的关系，并给出你对模型设定的结论；3）依据测算，选择你认为更能描述客观实际的模型，并简要说明其理由；4）根据分析结果，你对提高个人所得税起征点影响居民消费的有效性能得出什么结论?

7、练习题8.2参考解答： 录入如下数据obsCONSUMEINCOMEEMPLOYMENTD1D2D3D41985673.2000739.10002.1500000.0000000.0000000.0000000.0000001986799.0000900.90002.1200000.0000000.0000000.0000000.0000001987884.40001002.1002.0900001.0000000.0000000.0000000.00000019881104.0001180.2002.0300001.0000000.0000000.0000000.00000019891211

8、.0001373.9002.0000001.0000000.0000000.0000000.00000019901278.8901510.1601.9800001.0000000.0000000.0000000.00000019911453.8001700.6001.9600001.0000000.0000000.0000000.00000019921671.7002026.6001.9500001.0000000.0000000.0000000.00000019932110.8002577.4001.9200001.0000000.0000000.0000000.00000019942851

9、.3003496.2001.8800001.0000001.0000000.0000000.00000019953537.5704282.9501.8700001.0000001.0000000.0000000.00000019963919.5004838.9001.8600001.0000001.0000000.0000000.00000019974185.6005160.3001.8300001.0000001.0000000.0000000.00000019984331.6005425.1001.8000001.0000001.0000000.0000000.00000019994615

10、.9005854.0001.7700001.0000001.0000000.0000000.00000020004998.0006279.9801.6800001.0000001.0000000.0000000.00000020015309.0106859.6001.6500001.0000001.0000000.0000000.00000020026029.9207702.8001.5800001.0000001.0000000.0000000.00000020036510.9408472.2001.5800001.0000001.0000000.0000000.00000020047182

11、.1009421.6001.5600001.0000001.0000000.0000000.00000020057942.88010493.001.5100001.0000001.0000000.0000000.00000020068696.55011759.451.5300001.0000001.0000001.0000000.00000020079997.47013785.811.5400001.0000001.0000001.0000000.000000200811242.9015780.801.4800001.0000001.0000001.0000001.000000分别作如下回归：

12、Dependent Variable: CONSUMEMethod: Least SquaresDate: 08/24/09 Time: 13:14Sample (adjusted): 1986 2008Included observations: 23 after adjustmentsVariableCoefficientStd. Errort-StatisticProb.C744.7966378.06621.9700170.0676CONSUME(-1)0.0848730.0509071.6672210.1162INCOME0.6331180.03519817.987290.0000LO

13、G(EMPLOYMENT)-762.9720478.5280-1.5944140.1317D137.4346050.234450.7451980.4677D2221.076538.308405.7709660.0000D3-122.049373.81439-1.6534610.1190D4-178.868865.87071-2.7154520.0160R-squared0.999861Mean dependent var4428.906Adjusted R-squared0.999796S.D. dependent var3060.917S.E. of regression43.70477Ak

14、aike info criterion10.66100Sum squared resid28651.61Schwarz criterion11.05595Log likelihood-114.6015F-statistic15413.79Durbin-Watson stat2.977604Prob(F-statistic)0.000000Dependent Variable: CONSUMEMethod: Least SquaresDate: 08/24/09 Time: 13:14Sample (adjusted): 1986 2008Included observations: 23 af

15、ter adjustmentsVariableCoefficientStd. Errort-StatisticProb.C871.9310332.66272.6210670.0185CONSUME(-1)0.0835760.0501651.6660170.1152INCOME0.6299220.03444718.286760.0000LOG(EMPLOYMENT)-889.4616441.1508-2.0162300.0609D2226.036137.197916.0765790.0000D3-110.888471.26752-1.5559460.1393D4-171.692464.25105

16、-2.6722110.0167R-squared0.999856Mean dependent var4428.906Adjusted R-squared0.999802S.D. dependent var3060.917S.E. of regression43.09316Akaike info criterion10.61040Sum squared resid29712.33Schwarz criterion10.95598Log likelihood-115.0196F-statistic18496.74Durbin-Watson stat2.787479Prob(F-statistic)

17、0.000000Dependent Variable: CONSUMEMethod: Least SquaresDate: 08/24/09 Time: 13:15Sample (adjusted): 1986 2008Included observations: 23 after adjustmentsVariableCoefficientStd. Errort-StatisticProb.C1204.936265.10544.5451220.0003CONSUME(-1)0.0993140.0511471.9417090.0689INCOME0.5991650.02936620.40320

18、0.0000LOG(EMPLOYMENT)-1325.942354.4143-3.7412220.0016D2251.367534.815737.2199400.0000D4-141.771063.81647-2.2215430.0402R-squared0.999834Mean dependent var4428.906Adjusted R-squared0.999785S.D. dependent var3060.917S.E. of regression44.85802Akaike info criterion10.66434Sum squared resid34208.12Schwar

19、z criterion10.96056Log likelihood-116.6399F-statistic20483.46Durbin-Watson stat2.638666Prob(F-statistic)0.000000Dependent Variable: CONSUMEMethod: Least SquaresDate: 08/24/09 Time: 13:16Sample: 1985 2008Included observations: 24VariableCoefficientStd. Errort-StatisticProb.C1460.937233.29226.2622630.

20、0000INCOME0.6531010.00913271.515390.0000LOG(EMPLOYMENT)-1651.937314.1501-5.2584310.0000D2277.404833.627838.2492610.0000D4-154.274266.05969-2.3353770.0306R-squared0.999810Mean dependent var4272.418Adjusted R-squared0.999769S.D. dependent var3090.239S.E. of regression46.92598Akaike info criterion10.71

21、807Sum squared resid41838.91Schwarz criterion10.96350Log likelihood-123.6169F-statistic24931.15Durbin-Watson stat2.292463Prob(F-statistic)0.0000008.3 在统计学教材中，采用了方差分析方法分析了不同班次对劳动效率的影响，其样本数据为 表8.6 不同班次的劳动效率早班中班晚班374740355142334839335041355142365140374740试采用虚拟解释变量回归的方法对上述数据进行方差分析。练习题8.3参考解答：考虑到班次有三个属性，

22、故在有截距项的回归方程中只能引入两个虚拟变量，按加法形式引入，模型设定形式为： 其中，为劳动效率。 在Eviews中按下列格式录入数据： obsYD1D2134.000001.0000000.000000237.000001.0000000.000000335.000001.0000000.000000433.000001.0000000.000000533.000001.0000000.000000635.000001.0000000.000000736.000001.0000000.000000849.000000.0000001.000000947.000000.0000001.0000

23、001051.000000.0000001.0000001148.000000.0000001.0000001250.000000.0000001.0000001351.000000.0000001.0000001451.000000.0000001.0000001539.000000.0000000.0000001640.000000.0000000.0000001742.000000.0000000.0000001839.000000.0000000.0000001941.000000.0000000.0000002042.000000.0000000.0000002140.000000.

24、0000000.000000输入命令：ls y c d1 d2，则有如下结果Dependent Variable: YMethod: Least SquaresDate: 06/29/09 Time: 16:56Sample: 1 21Included observations: 21VariableCoefficientStd. Errort-StatisticProb.C40.428570.55532972.801150.0000D1-5.7142860.785353-7.2760690.0000D29.1428570.78535311.641710.0000R-squared0.9529

25、09Mean dependent var41.57143Adjusted R-squared0.947676S.D. dependent var*6.423172S.E. of regression1.469262Akaike info criterion3.738961Sum squared resid*38.85714Schwarz criterion3.888178Log likelihood-36.25909F-statistic*182.1176Durbin-Watson stat2.331933Prob(F-statistic)0.000000表中的*号部分表示在方差分析中需要用到的数据。依据上述数据，有：， 于是方差分析的结果为方差来源离差平方和自由度方差F值组间 786.2862393.143182.118组内 38.857182.158总