計量經(jīng)濟學(xué)第三版龐浩版課后答案

1、第二章2.2對于浙江省預(yù)算收入與全省生產(chǎn)總值的模型，用Eviews分析結(jié)果如下:R-squared0.983702Adjusted R-squared S.E.0.983177of regressionSum squared175.2325resid951899.7Log likelihood-216.275F-statistic1871.115由上可知，模型的參數(shù)：斜率系數(shù)0.176124,截品巨為一 154.3063Dependent Variable: YMethod: Least SquaresDate: 12/03/14 Time: 17:00Sample (adjusted): 1

2、 33Included observations: 33 after adjustmentsCoefficieVariable nt Std. Error t-Statistic Prob.X 0.176124 0.004072 43.25639 0.0000C3 39.08196 -3.9482740.0004Mean dependent902.514var8S.D. dependent1351.00var9Akaike info13.2288criterion013.3194 Schwarz criterion9Hannan-Quinn13.25931 criter.1Durbin-Wat

3、son0.10002stat1Prob(F-statistic) 0.000000關(guān)于浙江省財政預(yù)算收入與全省生產(chǎn)總值的模型，檢驗?zāi)Ｐ偷娘@著性：1）可決系數(shù)為0.983702 ,說明所建模型整體上對樣本數(shù)據(jù)擬合較好。2）對于回歸系數(shù)的t檢驗：t（b 2） =43.25639>t0.025 （31）=2.0395 ,對斜率系數(shù)的顯著性檢驗表明，全省生產(chǎn)總值對財政預(yù)算總收入有顯著影響。用規(guī)范形式寫出檢驗結(jié)果如下:Y=0.176124X 154.3063對于浙江省預(yù)算收入對數(shù)與全省生產(chǎn)總值對數(shù)的模型，由 Eviews分析結(jié)果如下:(0.004072)(39.08196)t= (43.25639

4、)(-3.948274 )R2=0.983702F=1871.115n=33經(jīng)濟意義是：全省生產(chǎn)總值每增加1億元，財政預(yù)算總收入增加0.176124億元(2)當(dāng) x=32000 時，進(jìn)行點預(yù)測，由上可知Y=0.176124X -154.3063 ,代入可得:Y= Y=0.176124*32000 154.3063=5481.6617Ex2=E (X X) =&x(n1)=進(jìn)行區(qū)間預(yù)測：先由Eviews分析:XYMean6000.441902.5148Median2689.280209.3900Maximum27722.314895.410Minimum123.720025.87000S

5、td. Dev.7608.0211351.009Skewness1.4325191.663108Kurtosis4.0105154.590432Jarque-Ber a12.6906818.69063Probability0.0017550.000087Sum198014.529782.99Sum Sq.Dev.1.85E+09 ；58407195Observation s3333由上7608.021 2 x (33 1)=1852223.473(Xf X)2=(32000 6000.441)2=675977068.2當(dāng)Xf=32000時，將相關(guān)數(shù)據(jù)代入計算得到：即 Yf 的置信區(qū)間為(548

6、1.6617 64.9649, 5481.6617+64.9649 )Dependent Variable: YMethod: Least SquaresDate: 12/03/14 Time: 18:00Sample (adjusted): 1 33Included observations: 33 after adjustmentsCoefficieProb.Variablent Std. Error t-StatisticLNX0.980275 0.034296 28.58268-1.918280.0000C9 0.268213 -7.1521210.0000Mean dependent

7、5.57312R-squared0.963442 var0AdjustedS.D.dependent1.68418R-squared0.962263 var9S.E. ofAkaike info0.66202regression0.327172 criterion8Sum squared0.75272resid3.318281 Schwarz criterion6-8.92346 Hannan-Quinn0.69254Log likelihood8 criter.5Durbin-Watson0.09620F-statistic816.9699 stat8Prob(F-statistic)0.0

8、00000模型方程為：lnY= 0.980275lnX-1.918289由上可知，模型的參數(shù)：斜率系數(shù)為0.980275 ,截距為-1.918289關(guān)于浙江省財政預(yù)算收入與全省生產(chǎn)總值的模型，檢驗其顯著性：1)可決系數(shù)為0.963442 ,說明所建模型整體上對樣本數(shù)據(jù)擬合較好。2)對于回歸系數(shù)的t檢驗：t ( ? =28.58268>t 0.025(31)=2.0395 ,對斜率系數(shù)的顯著性檢驗表明，全省生產(chǎn)總值對財政預(yù)算總收入有顯著影響。經(jīng)濟意義：全省生產(chǎn)總值每增長1%,財政預(yù)算總收入增長0.980275%2.4Method: Least SquaresDate: 12/01/14 T

9、ime: 12:40Sample: 1 12Included observations: 12CoefficieVariablent Std. Error t-StatisticProb.-64.1840X0 4.809828 -13.344340.0000C1845.475 19.26446 95.796880.0000Mean dependent1619.33R-squared0.946829 var3AdjustedS.D.dependent131.225R-squared0.941512 var2S.E. ofAkaike info9.90379regression31.73600 c

10、riterion2Sum squared9.98461resid10071.74 Schwarz criterion0-57.4227 Hannan-Quinn9.87387Log likelihood5 criter.1Durbin-Watson1.17240F-statistic178.0715 stat7Prob(F-statistic)0.000000由上可得：建筑面積與建造成本的回歸方程為：Y=1845.475-64.18400X經(jīng)濟意義：建筑面積每增加1萬平方米，建筑單位成本每平方米減少64.18400Dependent Variable: Y首先進(jìn)行點預(yù)測，由 Y=1845.47

11、5-64.18400X 得,當(dāng) x=4.5 , y=1556.647再進(jìn)行區(qū)間估計:用Eviews分析:YXMean1619.3333.523333Median1630.0003.715000Maximum1860.0006.230000Minimum1419.0000.600000Std. Dev.131.22521.989419Skewness0.003403-0.060130Kurtosis2.3465111.664917Jarque-Ber a0.2135470.898454Probability0.8987290.638121Sum19432.0042.28000Sum Sq. De

12、v.189420.743.53567Observation s1212由上表可知Ex2=E (XX) 2=6 2x(n1)=1.989419 2 x (12 1)=43.5357 (X f X)2=(4.5 3.523333) 2=0.95387843當(dāng)Xf=4.5時，將相關(guān)數(shù)據(jù)代入計算得至U :1556.647 2.228x 31.73600 x , 1/12+43.5357/0.95387843<Yf< 1556.647 +2.228x31.73600 x , 1/12+43.5357/0.95387843即 Yf 的置信區(qū)間為( 1556.647 478.1231, 1556

13、.647 +478.1231 )3.2Method: Least SquaresDate: 12/01/14 Time: 20:25Sample: 1994 2011Included observations: 18CoefficieProb.Variablent Std. Error t-StatisticX20.135474 0.012799 10.584540.0000X318.85348 9.776181 1.9285120.0729-18231.5C8 8638.216 -2.1105730.0520Mean dependent6619.19R-squared0.985838 var

14、1AdjustedS.D.dependent5767.15R-squared0.983950 var2S.E. ofAkaike info16.1767regression730.6306 criterion0Sum squared16.3251resid8007316. Schwarz criterion0-142.590 Hannan-Quinn16.1971Log likelihood3 criter.7Durbin-Watson1.17343F-statistic522.0976 stat2Prob(F-statistic)0.000000由上可知，模型為：Y = 0.135474X

15、2 + 18.85348X 3 - 18231.58對模型進(jìn)行檢驗：1)可決系數(shù)是0.985838 ,修正的可決系數(shù)為0.983950 ,說明模型對樣本擬合較好2) F 檢驗，F(xiàn)=522.0976>F (2,15)=4.77,回歸方程顯著3) t檢驗，t統(tǒng)計量分別為X2的系數(shù)對應(yīng)t值為10.58454 ,大于t (15) =2.131 , 系數(shù)是顯著的，X3的系數(shù)對應(yīng)t值為1.928512，小于t (15) =2.131 ,說明此系數(shù)是不顯著的(2)對于對數(shù)模型，用Eviews分析結(jié)果如下：Dependent Variable: LNYMethod: Least SquaresDate:

16、 12/01/14 Time: 20:25Sample: 1994 2011Included observations: 18VariableCoefficient Std. Error t-StatisticProb.0.00000.0209LNX21.564221 0.088988 17.57789LNX31.760695 0.682115 2.581229-20.52048 5.432487 -3.7773630.0018Mean dependent8.40011R-squared0.986295 var2AdjustedS.D. dependent0.94153R-squared0.9

17、84467 var0S.E. ofAkaike info-1.2964regression0.117343 criterion24Sum squared0.206540 Schwarz criterion”48029Hannan-Quinn-1.2759Log likelihood14.66782 criter.62F-statisticDurbin-Watson0.68665539.7364 statProb(F-statistic)0.000000由上可知，模型為：LNY=-20.52048+1.564221 LNX 2+1.760695 LNX 3對模型進(jìn)行檢驗：1)可決系數(shù)是0.986

18、295，修正的可決系數(shù)為0.984467 ,說明模型對樣本擬合較好。2) F 檢驗，F(xiàn)=539.7364> F (2,15) =4.77,回歸方程顯著。3) t檢驗，t統(tǒng)計量分別為-3.777363,17.57789,2.581229 , 均大于t (15) =2.131 , 所 以這些系數(shù)都 是顯著的。(3)(1)式中的經(jīng)濟意義：工業(yè)增加 1億元，出口貨物總額增加0.135474億元，人民幣匯 率增加1,出口貨物總額增加18.85348億元。(2)式中的經(jīng)濟意義：工業(yè)增加額每增加1%,出口貨物總額增加1.564221%，人民幣匯率每增加1%,出口貨物總額增加1.760695%3.3對家

19、庭書刊消費對家庭月平均收入和戶主受教育年數(shù)計量模型，由 Eviews分析結(jié)果如下：Dependent Variable: YMethod: Least SquaresDate: 12/01/14 Time: 20:30Sample: 1 18R-squared AdjustedR-squaredS.E. of regression Sum squared residMean dependent 0.951235 varS.D.dependent 0.944732 varAkaike info 60.82273 criterion55491.07 Schwarz criterionLog lik

20、elihoodF-statistic-97.8433 Hannan-Quinn 4 criter.Durbin-Watson146.2974 stat755.1222258.720611.2048211.3532111.225282.605783Included observations: 18CoefficieProb.Variablent Std. Error t-StatisticX0.086450 0.029363 2.9441860.0101T52.37031 5.202167 10.06702-50.01630.0000C8 49.46026 -1.0112440.327911.5

21、6343對殘差進(jìn)行模型分析，用 Eviews分析結(jié)果如下：模型為：Y = 0.086450X + 52.37031T-50.01638對模型進(jìn)行檢驗：1)可決系數(shù)是0.951235，修正的可決系數(shù)為0.944732 ,說明模型對樣本擬合較好。2) F 檢驗，F(xiàn)=539.7364> F (2,15) =4.77,回歸方程顯著。3) t檢驗，t統(tǒng)計量分別為2.944186,10.06702 , 均大于t (15) =2.131 ,所以這些系數(shù)都是顯著的。經(jīng)濟意義：家庭月平均收入增加1元，家庭書刊年消費支出增加0.086450元，戶主受教育年數(shù)增加1年，家庭書刊年消費支出增加52.37031元

22、。用Eviews分析：Dependent Variable: YMethod: Least SquaresDate: 12/01/14 Time: 22:30Sample: 1 18Included observations: 18Variablent Std. Error t-Statistic Prob.T 63.01676 4.548581 13.85416 0.0000C1 58.02290 -0.1996060.8443Mean dependent755.122R-squared Adjusted R- squared S.E. of regression Sum squared r

23、esid0.923054 var2S.D. dependent258.7200.918245 var6Akaike info11.549773.97565 criterion987558.36 Schwarz criterionLog likelihoodF-statistic-101.948 Hannan-Quinn 1criter.Durbin-Watson 2.13404191.9377 statProb(F-statistic)0.000000Dependent Variable: XMethod: Least SquaresDate: 12/01/14 Time: 22:34Samp

24、le: 1 18Included observations: 18Coefficient Std. Error t-Statistic Prob.VariableT123.151631.841503.8676440.0014C444.5888406.17861.0945650.2899Mean dependent1942.93R-squared0.483182 var3AdjustedS.D. dependent698.832R-squared0.450881 var5S.E. ofAkaike info15.4417regression517.8529 criterion0Sum squar

25、ed15.5406resid4290746. Schwarz criterion3-136.975 Hannan-Quinn15.4553Log likelihood3 criter.4Durbin-Watson1.05225F-statistic14.95867 stat1Prob(F-statistic) 0.001364以上分別是y與T, X與T的一元回歸模型分別是：Y = 63.01676T - 11.58171X = 123.1516T + 444.5888Method: Least SquaresMethod: Least SquaresDate: 12/03/14 Time: 2

26、0:39Sample: 1 18VariableIncluded observations: 18nt Std. Error t-Statistic Prob.E20.086450 0.028431 3.040742 0.0078C3.96E-14 13.88083 2.85E-15 1.0000Mean dependent2.30E-1R-squared Adjusted R- squared S.E. of regression Sum squared resid0.366239 var4S.D.dependent71.76690.326629 var3Akaike info11.0937

27、58.89136 criterion055491.07 Schwarz criterion 4Log likelihood-97.8433 Hannan-Quinn11.1073F-statisticDurbin-Watson2.605784 criter.9.246111 stat3Prob(F-statistic) 0.007788模型為：日=0.086450E 2 + 3.96e-14參數(shù)：斜率系數(shù)a為0.086450,截距為3.96e-14由上可知，B2與a2的系數(shù)是一樣的。回歸系數(shù)與被解釋變量的殘差系數(shù)是一樣的它們的變化規(guī)律是一致的。第五章5.3 (1)由Eviews軟件分析得：Da

28、te: 12/10/14 Time: 16:00Sample: 1 31Included observations: 31CoefficieProb.Variablent Std. Error t-StatisticX1.244281 0.079032 15.744110.0000C242.4488 291.1940 0.8326020.4119Mean dependent4443.52R-squared0.895260 var6AdjustedS.D.dependent1972.07R-squared0.891649 var2S.E. ofAkaike info15.8515regressi

29、on649.1426 criterion2Sum squared122201915.9440resid6 Schwarz criterion4-243.698 Hannan-Quinn15.8816Log likelihood6 criter.8Durbin-Watson1.07858F-statistic247.8769 stat1Prob(F-statistic)0.000000由上表可知，2007年我國農(nóng)村居民家庭人均消費支出（x）對人均純收入（y）的模型為:Y=1.244281X+242.4488(2)由圖形法檢驗由上圖可知，模型可能存在異方差 Goldfeld-Quanadt 檢驗1

30、）定義區(qū)間為1-12時，由軟件分析得Dependent Variable: Y1Method: Least SquaresSample: 1 12Included observations: 12Variablent Std. Error t-Statistic Prob.X11.485296 0.500386 2.9682970.0141-550.549C2 1220.063 -0.4512470.6614Mean dependent3052.95R-squared0.468390var0AdjustedS.D. dependent550.514R-squared0.415229var8S.

31、E. ofAkaike info15.0740regression420.9803criterion6Sum squared15.1548resid1772245. Schwarz criterion8-88.4443 Hannan-Quinn15.0441Log likelihood7 criter.4Durbin-Watson2.35416F-statistic8.810789 stat7Prob(F-statistic)0.014087得三 e1i 2=1772245.2)定義區(qū)間為20-31時，由軟件分析得Dependent Variable: Y1Method: Least Squa

32、resDate: 12/10/14 Time: 16:36Sample: 20 31Included observations: 12CoefficieVariablent Std. Errort-StatisticProb.X11.0869400.1488637.3016230.0000C1173.307733.25201.6001410.1407Mean dependent6188.32R-squared0.842056Adjusted R- squared S.E.0.826262of regression Sum squared889.3633resid 7909670.Log lik

33、elihood_97.4194F-statisticvar9S.D. dependent2133.69var2Akaike info16.5699criterion016.6507Schwarz criterion2Hannan-Quinn16.53990 criter.8Durbin-Watson2.3397653.31370 stat7Prob(F-statistic) 0.000026得匯 e2i 2=7909670.3)根據(jù)Goldfeld-Quanadt 檢驗，F(xiàn)統(tǒng)計量為:F4e2i2 / Ee 1i2 =7909670./ 1772245=4.4631在a =0.05水平下，分子分

34、母的自由度均為10,查分布表得臨界值F0.05 (10,10) =2.98,因為F=4.4631> F 0.05 ( 10,10 ) =2.98, 所以拒絕原假設(shè)，此檢驗表明模型存在異方差。(3)1)采用WL嗨估計過程中，用權(quán)數(shù)w1=1/X,建立回歸得：Dependent Variable: YMethod: Least SquaresDate: 12/09/14 Time: 11:13Sample: 1 31Included observations: 31Weighting series: W1Variablent Std. Error t-Statistic Prob.X1.425

35、859 0.119104 11.971570.00000.3389C-334.813 344.3523 -0.9722981Dependent Variable: WGT_RESID A2Weighted StatisticsR-squared0.831707Adjusted R- squared S.E.0.825904of regression Sum squared536.6796resid 8352726.Log likelihood-237.800F-statistic 143.3184Prob(F-statistic) 0.000000Unweighted StatisticsMe

36、an dependent varS.D.dependent varAkaike info criterionSchwarz criterionHannan-Quinn 8 criter.Durbin-Watson stat3946.082536.190715.4710215.5635415.501181.369081R-squared0.875855AdjustedR-squared0.871574S.E. of regression706.7236Durbin-Watson stat1.532908Mean dependent 4443.52 varS.D.dependent varSum

37、squared resid61972.07214484289對此模型進(jìn)行White檢驗得：Heteroskedasticity Test: WhiteF-statistic0.299395 Prob. F(2,28) Prob.0.649065 Chi-Square(2) Prob.Obs*R-squared1.798067 Chi-Square(2)Scaled explainedSS0.74360.72290.4070Test Equation:Method: Least SquaresDate: 12/10/14 Time: 21:13Sample: 1 31Included obser

38、vations: 31VariableCollinear test regressors dropped from specificationnt Std. Error t-Statistic Prob.C61927.89 1045682. 0.059222 0.9532WGA29 1173622. -0.506064 0.6168X*WGA2282.4407 747.9780 0.377606 0.7086Mean 269442.R-squareddependent 0.020938 var -0.04899 S.D.8Adjusted R- dependent 5 var689166.sq

39、uared S.E.Akaike info 705847.65of regressioncriterion29.8639Sum squared 1.40E+1 3 Schwarz criterion -459.8915residHannan-Quinn30.00273 criter.3Log likelihoodDurbin-Watson 0.29939529.9091stat9F-statistic1.92233Prob(F-statistic) 0.7436106Method: Least SquaresDate: 12/09/14Sample: 1 31Time: 21:08Includ

40、ed observations: 31Weighting series: W2CoefficieProb.Variablent Std. Error t-StatisticX1.557040 0.145392 10.709220.0000-693.194C6 376.4760 -1.8412720.0758Weighted StatisticsMean dependent3635.02R-squared0.798173 var8AdjustedS.D.dependent1029.83R-squared0.791214 var0S.E. ofAkaike info15.1922regressio

41、n466.8513 criterion4Sum squared15.2847resid6320554. Schwarz criterion5-233.479 Hannan-Quinn15.2224Log likelihood7 criter.0Durbin-Watson1.56297F-statistic114.6875 stat5Prob(F-statistic)0.000000UnweightedStatisticsMean dependent4443.52R-squared0.834850 var6AdjustedS.D.dependent1972.07R-squared0.829156

42、 var2S.E. ofSum squared192683regression815.1229 resid34Durbin-Watsonstat1.678365對此模型進(jìn)行White檢驗得:Heteroskedasticity Test: WhiteProb(F-statistic)0.825233F-statisticObs*R-squaredScaled explainedSS0.299790Prob. f(3,27)Prob.0.999322Chi-Square(3)Prob.1.789507Chi-Square(3)0.82520.80140.6172Test Equation:Dep

43、endent Variable: WGT_RESID A2Method: Least SquaresDate: 12/10/14 Time: 21:29Sample: 1 31VariableIncluded observations: 31nt Std. Error t-Statistic Prob.-111661.549855.7 -0.2030750.8406WGTA2XA2*WGTA2426220.22240181.0.1902620.1948880.516395 0.3774020.85050.7088X*WGTA2R-squared AdjustedR-squaredS.E. of

44、regression Sum squared residLog likelihoodF-statistic0.7816-583.2151 2082.820 -0.280012Mean dependent203888.0.032236 var8-0.07529 S.D. dependent419282.3 var0Akaike info28.9229434780.1 criterion85.10E+129.10802 Schwarz criterion1-444.306 Hannan-Quinn28.98332 criter.0Durbin-Watson1.835850.299790 stat4

45、nR2=0.999322V JT? 0.05從上可知，nR2=0.999322 ,比較計算的/統(tǒng)計量的臨界值，因為=5.9915 ,所以接受原假設(shè)，該模型消除了異方差。估計結(jié)果為：Y=1.557040X-693.1946t= (10.70922 ) (-1.841272 ) R2=0.798173 F=114.6875 DW=1.562975用權(quán)數(shù)w3=1/sqr (x),用回歸分析得：Dependent Variable: YMethod: Least SquaresDate: 12/09/14 Time: 21:35Sample: 1 31Included observations: 31

46、Coefficient Std. Error t-StatisticWeighting series: W3X1.330130 0.098345 13.525070.0000-47.4024C2 313.1154 -0.1513900.8807Weighted StatisticsMean dependent4164.11R-squared0.863161 var8AdjustedS.D.dependent991.207R-squared0.858442 var9S.E. ofAkaike info15.6501regression586.9555 criterion2Sum squared1

47、5.7426resid9990985. Schwarz criterion3-240.576 Hannan-Quinn15.6802Log likelihood8 criter.7Durbin-Watson1.23766F-statistic182.9276 stat4Prob(F-statistic)0.000000VariableProb.UnweightedStatisticsR-squared0.890999Adjusted R- squared0.887240S.E. of regression662.2171Durbin-Watson stat1.314859Mean depend

48、ent var4443.52S.D. dependent var6Sum squared resid1972.07212717412對此模型進(jìn)行White檢驗得:Heteroskedasticity Test: White0.423886 Prob. F(2,28) Prob.0.911022 Chi-Square(2) Prob.2.768332 Chi-Square(2)0.65860.63410.2505F-statisticObs*R-squaredScaled explainedSSTest Equation:Dependent Variable: WGT_RESID A2Metho

49、d: Least SquaresDate: 12/09/14 Time: 20:36Sample: 1 31Included observations: 31Collinear test regressors dropped from specificationVariableCoefficient Std. Errort-Statistic Prob.1212308.2141958.0.565981WGTA2XA2*WGTA2-715673.-0.0151901301839.4 0.082276-0.549740-0.1846770.57590.58690.8548R-squared0.02

50、9388 Mean dependent 322289.varAdjusted R- squared S.E. of regression Sum squared residLog likelihoodF-statistic-0.03994 S.D. dependent 863356.2 var7Akaike info30.3059880429.8 criterion72 7E+! Schwarz criterion. 444-466.742 Hannan-Quinn30.3512 6 criter.1Durbin-Watson1.887420.423886 stat6Prob(F-statistic) 0.658628從上可知，nR2=0.911022 ,比較計算的/統(tǒng)計量的臨界值，因為nR=0.911022<*'0.05=5.9915 ,所以接受原假設(shè)，該模型消除了異方差。估計結(jié)果為：Y=1.330130X-47.40242t= (13.52507 ) (-0.151390 ) R2=0.863161 F=182.9276 DW=1.237664經(jīng)過檢驗發(fā)現(xiàn)，用權(quán)數(shù) w1的效果最好，