【正文】 ed. dependent var. of regressionSchwarz criterionLog likelihoodDurbinWatson statProb(Fstatistic)得∑e2i2=7909670.3）根據(jù)GoldfeldQuanadt檢驗，F(xiàn)統(tǒng)計量為：F=∑e2i2 /∑e1i2 =7909670./ 1772245=在α=，分子分母的自由度均為10，（10,10）=，因為F= （10,10）=，所以拒絕原假設，此檢驗表明模型存在異方差。XCWeighted StatisticsRsquared. dependent var. of regressionSchwarz criterionLog likelihoodDurbinWatson statProb(Fstatistic)Unweighted StatisticsRsquared. dependent var. of regressionProb. F(3,27)Obs*RsquaredProb. ChiSquare(3)Test Equation:Dependent Variable: WGT_RESID^2Method: Least SquaresDate: 12/24/15 Time: 16:45Sample: 1 31Included observations: 31VariableCoefficientStd. ErrortStatisticProb.估計結果為： Y= t=（）（）R2= F= DW=經(jīng)過檢驗發(fā)現(xiàn)，用權數(shù)w1的效果最好，所以綜上可知，即修改后的結果為：Y= t=（）（）R2= F= DW=(1)a)用Eviews模型分析得：Dependent Variable: YMethod: Least SquaresDate: 12/24/15 Time: 19:16Sample: 1978 2011Included observations: 34VariableCoefficientStd. ErrortStatisticProb.Mean dependent varAdjusted RsquaredAkaike info criterionSum squared residHannanQuinn criter.FstatisticDurbinWatson statProb(Fstatistic)得∑e2i2=3）根據(jù)GoldfeldQuanadt檢驗，F(xiàn)統(tǒng)計量為：F=∑e2i2 /∑e1i2 =在α=，分子分母的自由度均為11，（11,11）=，因為F= （11,11）=，所以拒絕原假設，此檢驗表明模型存在異方差。Schwarz criterionLog likelihood. dependent var. of regressionXCRsquaredHannanQuinn criter.FstatisticAkaike info criterionSum squared residMean dependent varAdjusted RsquaredHannanQuinn criter.FstatisticAkaike info criterionSum squared resid+12Mean dependent varAdjusted Rsquared估計結果為： Y= t=（）（）R2= F= DW=②用權數(shù)w2=1/x2,用回歸分析得：Dependent Variable: YMethod: Least SquaresDate: 12/24/15 Time: 16:40Sample: 1 31Included observations: 31Weighting series: W2VariableCoefficientStd. ErrortStatisticProb.DurbinWatson statProb(Fstatistic)得∑e1i2=1772245.2）定義區(qū)間為2031時，由軟件分析得：Dependent Variable: Y1Method: Least SquaresDate: 12/24/15 Time: 16:16Sample: 20 31Included observations: 12VariableCoefficientStd. ErrortStatisticProb.Schwarz criterionLog likelihood. dependent var. of regressionX1CRsquaredDurbinWatson statProb(Fstatistic)得∑e2i2=3）根據(jù)GoldfeldQuanadt檢驗，F(xiàn)統(tǒng)計量為：F=∑e2i2 /∑e1i2 =在α=，分子分母的自由度均為4，（4,4）=，因為F= （4,4）=，所以接受原假設，此檢驗表明模型不存在異方差。Schwarz criterionLog likelihood. dependent var. of regressionTXCRsquared由上表可知，CZZC與GDP，CZZC與SSZE，GDP與SSZE之間的相關系數(shù)都非常高，說明確實存在多重共線性。GDP（2）得到相關系數(shù)矩陣如下：CZSRCZZCGDPSSZECZSRHannanQuinn criter.FstatisticAkaike info criterionSum squared resid2866884.Mean dependent varAdjusted RsquaredDurbinWatson statProb(Fstatistic)①得到的回歸方程分別為1）LNY= 2）LNY= 3）LNGDPt= ②對多重共線性的認識：單方程擬合效果都很好，回歸系數(shù)顯著，判定系數(shù)較高，GDP和CPI對進口的顯著的單一影響，在這兩個變量同時引入模型時影響方向發(fā)生了改變，這只有通過相關系數(shù)的分析才能發(fā)現(xiàn)。Schwarz criterionLog likelihood. dependent var. of regressionLNCPICRsquaredHannanQuinn criter.FstatisticAkaike info criterionSum squared residMean dependent varAdjusted Rsquared②得到相關系數(shù)矩陣如下：LNYLNGDPLNCPILNYHannanQuinn criter.FstatisticAkaike info criterionSum squared residMean dependent varAdjusted RsquaredDurbinWatson statProb(Fstatistic)模型為：E1 = + 參數(shù)：，（3）由上可知，β2與α2的系數(shù)是一樣的。Schwarz criterionLog likelihood. dependent var. of regressionE2CRsquaredHannanQuinn criter.FstatisticAkaike info criterionSum squared residMean dependent varAdjusted Rsquared3）t檢驗，均大于t（15）=，所以這些系數(shù)都是顯著的。DurbinWatson statProb(Fstatistic)①由上可知，模型為：LNY=+ LNX2+ LNX3②對模型進行檢驗：1），說明模型對樣本擬合較好。Schwarz criterionLog likelihood. dependent var. of regressionLNX2LNX3CRsquared（3）用EViews分析得：Dependent Variable: YMethod: Least SquaresDate: 12/23/15 Time: 21:09Sample: 1 31Included observations: 31VariableCoefficientStd. ErrortStatisticProb. x (12—1)=(Xf—X)2=(—Median2）對于回歸系數(shù)的t檢驗：t（β2）=(31)=，對斜率系數(shù)的顯著性檢驗表明，全省生產總值對財政預算總收入有顯著影響。58407195ProbabilityMeanHannanQuinn criter.FstatisticAkaike info criterionSum squared residMean dependent varAdjusted Rsquared③關于人均壽命與一歲兒童疫苗的模型，由上可知，說明所建模型整體上對樣本數(shù)據(jù)擬合較好。HannanQuinn criter.FstatisticAkaike info criterionSum squared residMean dependent varAdjusted RsquaredDurbinWatson statProb(Fstatistic)有上可知，關系式為y=+②關于人均壽命與成人識字率的關系，用Eviews分析如下：Dependent Variable: YMethod: Least SquaresDate: 12/23/15 Time: 15:01Sample