計(jì)量經(jīng)濟(jì)學(xué)第三次作業(yè)

下表列出了某年中國(guó)部分省市城鎮(zhèn)居民每個(gè)家庭平均全年可支配收入X與消費(fèi)性支出Y的統(tǒng)計(jì)數(shù)據(jù)。地區(qū)可支配收入（X）消費(fèi)性支出（Y）地區(qū)可支配收入（X）消費(fèi)性支出（Y）北京10349.698493.49浙江9279.167020.22天津8140.506121.04山東6489.975022.00河北5661.164348.47河南4766.263830.71山西4724.113941.87湖北5524.544644.5內(nèi)蒙古5129.053927.75湖南6218.735218.79遼寧5357.794356.06廣東9761.578016.91吉林4810.004020.87陜西5124.244276.67黑龍江4912.883824.44甘肅4916.254126.47上海11718.018868.19青海5169.964185.73江蘇6800.235323.18新疆5644.864422.93(1)試用普通最小二乘法建立居民人均消費(fèi)支出與可支配收入的線性模型；(2)檢驗(yàn)?zāi)Ｐ褪欠翊嬖诋惙讲钚裕?3)如果存在異方差性，試采用適當(dāng)?shù)姆椒ü烙?jì)模型參數(shù)。解：（1）a.建立對(duì)象，錄入可支配收入X與消費(fèi)性支出Y，如下圖：b.設(shè)定一元線性回歸模型為：點(diǎn)擊主界面菜單Quick\EstimateEquation，在彈出的對(duì)話框中輸入Y、C、X，操作如下圖：（2）a.生成殘差序列。在工作文件中點(diǎn)擊Object\GenerateSeries?，在彈出的窗口中，在主窗口鍵入命令如下“e1=resid^2”得到殘差平方和序列e1b.繪制e1與x的散點(diǎn)圖。按住Ctrl鍵，同時(shí)選擇變量X與e2以組對(duì)象方式打開(kāi)，進(jìn)入數(shù)據(jù)列表，再點(diǎn)擊View\Graph\Scatter\SimpleScatter，可得散點(diǎn)圖。如上圖:(3)a.設(shè)定一元線性回歸模型為：點(diǎn)擊主界面菜單Quick\EstimateEquation，在彈出的對(duì)話框中輸入log(e1)、C、X，得出結(jié)果如下圖：b.在工作文件中點(diǎn)擊Object\GenerateSeries，在彈出的窗口中，在主窗口鍵入命令如下”w=1/sqr(exp(6.8251+0.00046*x))”得出權(quán)數(shù)W.c.點(diǎn)擊主界面菜單Quick\EstimateEquation，在彈出的Specification對(duì)話框中輸入Y、C、X，在Options中的Weightseries中填入權(quán)數(shù)w.如下圖：（1）結(jié)果DependentVariable:YMethod:LeastSquaresDate:12/5/16Time:19:57Sample:120Includedobservations:20VariableCoefficientStd.Errort-StatisticProb.

C272.3635159.67731.7057130.1053X0.7551250.02331632.386900.0000R-squared0.983129

Meandependentvar5199.515AdjustedR-squared0.982192

S.D.dependentvar1625.275S.E.ofregression216.8900

Akaikeinfocriterion13.69130Sumsquaredresid846743.0

Schwarzcriterion13.79087Loglikelihood-134.9130

Hannan-Quinncriter.13.71073F-statistic1048.912

Durbin-Watsonstat1.301684Prob(F-statistic)0.000000得到模型的估計(jì)結(jié)果為：(1.706)(32.387)F=1048.912估計(jì)結(jié)果顯示，即使在10%的顯著性水平下，都不拒絕常數(shù)項(xiàng)為零的假設(shè)。（2）由b的圖可知，殘差平方e1與x大致存在遞增關(guān)系，即存在單調(diào)增型異方差。（3）通過(guò)加權(quán)得出的方程結(jié)果如下：Method:LeastSquaresDate:12/5/16Time:20:01Sample:120Includedobservations:20Weightingseries:WWeighttype:Standarddeviation(averagescaling)HACstandarderrors&covariance(Bartlettkernel,Newey-Westfixed

bandwidth=3.0000)VariableCoefficientStd.Errort-StatisticProb.

C383.2138198.13011.9341520.0690X0.7420820.02400230.917420.0000WeightedStatisticsR-squared0.975308

Meandependentvar5232.501AdjustedR-squared0.973936

S.D.dependentvar5438.029S.E.ofregression296.8192

Akaikeinfocriterion14.31876Sumsquaredresid1585830.

Schwarzcriterion14.41834Loglikelihood-141.1876

Hannan-Quinncriter.14.33820F-statistic710.9713

Durbin-Watsonstat1.881095Prob(F-statistic)0.000000

Weightedmeandep.7229.207WaldF-statistic955.8868

Prob(WaldF-statistic)0.000000UnweightedStatisticsR-squared0.982571

Meandependentvar5199.515AdjustedR-squared0.981603

S.D.dependentv