計(jì)量經(jīng)濟(jì)學(xué)多元線性回歸、多重共線性、異方差實(shí)驗(yàn)報(bào)告

計(jì)量經(jīng)濟(jì)學(xué)多元線性回歸、多重共線性、異方差實(shí)驗(yàn)報(bào)告計(jì)量經(jīng)濟(jì)學(xué)實(shí)驗(yàn)報(bào)告多元線性回歸、多重共線性、異方差實(shí)驗(yàn)報(bào)告一、研究目的和要求：隨著經(jīng)濟(jì)的發(fā)展，人們生活水平的提高，旅游業(yè)已經(jīng)成為中國(guó)社會(huì)新的經(jīng)濟(jì)增長(zhǎng)點(diǎn)。旅游產(chǎn)業(yè)是一個(gè)關(guān)聯(lián)性很強(qiáng)的綜合產(chǎn)業(yè)，一次完整的旅游活動(dòng)包括吃、住、行、游、購(gòu)、娛六大要素，旅游產(chǎn)業(yè)的發(fā)展可以直接或者間接推動(dòng)第三產(chǎn)業(yè)、第二產(chǎn)業(yè)和第一產(chǎn)業(yè)的發(fā)展。尤其是假日旅游，有力刺激了居民消費(fèi)而拉動(dòng)內(nèi)需。2012年，我國(guó)全年國(guó)內(nèi)旅游人數(shù)達(dá)到30.0億人次，同比增長(zhǎng)13.6%，國(guó)內(nèi)旅游收入2.3萬(wàn)億元，同比增長(zhǎng)19.1%。旅游業(yè)的發(fā)展不僅對(duì)增加就業(yè)和擴(kuò)大內(nèi)需起到重要的推動(dòng)作用，優(yōu)化產(chǎn)業(yè)結(jié)構(gòu)，而且可以增加國(guó)家外匯收入，促進(jìn)國(guó)際收支平衡，加強(qiáng)國(guó)家、地區(qū)間的文化交流。為了研究影響旅游景區(qū)收入增長(zhǎng)的主要原因，分析旅游收入增長(zhǎng)規(guī)律，需要建立計(jì)量經(jīng)濟(jì)模型。影響旅游業(yè)發(fā)展的因素很多，但據(jù)分析主要因素可能有國(guó)內(nèi)和國(guó)際兩個(gè)方面，因此在進(jìn)行旅游景區(qū)收入分析模型設(shè)定時(shí)，引入城鎮(zhèn)居民可支配收入和旅游外匯收入為解釋變量。旅游業(yè)很大程度上受其產(chǎn)業(yè)本身的發(fā)展水平和從業(yè)人數(shù)影響，固定資產(chǎn)和從業(yè)人數(shù)體現(xiàn)了旅游產(chǎn)業(yè)發(fā)展規(guī)模的內(nèi)在影響因素，因此引入旅游景區(qū)固定資產(chǎn)和旅游業(yè)從業(yè)人數(shù)作為解釋變量。因此選取我國(guó)31個(gè)省市地區(qū)的旅游業(yè)相關(guān)數(shù)據(jù)進(jìn)行定量分析我國(guó)旅游業(yè)發(fā)展的影響因素。二、模型設(shè)定根據(jù)以上的分析，建立以下模型Y=β+β1X+β2X+βX+βX+Ut參數(shù)說明：Y——旅游景區(qū)營(yíng)業(yè)收入/萬(wàn)元X——旅游業(yè)從業(yè)人員/人X——旅游景區(qū)固定資產(chǎn)/萬(wàn)元青海638.4387419851.28265915603.31寧夏49509.861219623149.9062017578.92新疆28993.114045152280.364651915513.62數(shù)據(jù)來源：1.中國(guó)統(tǒng)計(jì)年鑒2012，2.中國(guó)旅游年鑒2012。三、參數(shù)估計(jì)利用Eviews6.0做多元線性回歸分析步驟如下：1、創(chuàng)建工作文件雙擊Eviews6.0圖標(biāo)，進(jìn)入其主頁(yè)。在主菜單中依次點(diǎn)擊“File\New\Workfile”，出現(xiàn)對(duì)話框“WorkfileRange”。本例中是截面數(shù)據(jù)，在workfilestructuretype中選擇“Unstructured/Undated”,在Daterange中填入observations31,點(diǎn)擊ok鍵，完成工作文件的創(chuàng)建。2、輸入數(shù)據(jù)在命令框中輸入dataYX1X2X3X4,回車出現(xiàn)“Group”窗口數(shù)據(jù)編輯框，在對(duì)應(yīng)的YX1X2X3X4下輸入相應(yīng)數(shù)據(jù)，關(guān)閉對(duì)話框?qū)⑵涿麨間roup01,點(diǎn)擊ok,保存。對(duì)數(shù)據(jù)進(jìn)行存盤，點(diǎn)擊“File/SaveAs”，出現(xiàn)“SaveAs”對(duì)話框，選擇存入路徑，并將文件命名，再點(diǎn)“ok”。3、參數(shù)估計(jì)在Eviews6.0命令框中鍵入“LSYCX1X2X3X4”利用Eviews6.0估計(jì)模型參數(shù)，最小二乘法的回歸結(jié)果如下：表3.1回歸結(jié)果DependentVariable:YMethod:LeastSquaresDate:11/14/13Time:21:14Sample:131Includedobservations:31CoefficientStd.Errort-StatisticProb.

C32390.8339569.490.8185810.4205X10.6036240.3661121.6487410.1112X20.2342650.0412185.6835830.0000X30.0446320.0607550.7346200.4691X4-1.9140342.098257-0.9122020.3700R-squared0.879720

Meandependentvar114619.2AdjustedR-squared0.861215

S.D.dependentvar112728.1S.E.ofregression41995.55

Akaikeinfocriterion24.27520Sumsquaredresid4.59E+10

Schwarzcriterion24.50649Loglikelihood-371.2657

Hannan-Quinncriter.24.35060F-statistic47.54049

Durbin-Watsonstat2.007191Prob(F-statistic)0.000000根據(jù)表中的樣本數(shù)據(jù)，模型估計(jì)結(jié)果為=32390.83+0.603624X+0.234265X+0.044632X-1.914034X（39569.49）（0.366112）（0.041218）（0.060755）（2.098257）t=(0.818581)(1.648741)(5.683583)(0.734620)(-0.912202)R2=0.879720=0.861215F=47.54049DW=2.007191可以看出，可決系數(shù)R2=0.879720，修正的可決系數(shù)=0.861215。說明模型的擬合程度還可以。但是當(dāng)α=0.05時(shí)，X、X、X系數(shù)均不能通過檢驗(yàn)，且X的系數(shù)為負(fù)，與經(jīng)濟(jì)意義不符，表明模型很可能存在嚴(yán)重的多重共線性。四、模型修正1．多重共線性的檢驗(yàn)與修正（1）檢驗(yàn)選中X1X2X3X4數(shù)據(jù)，點(diǎn)擊右鍵，選擇“Open/asGroup”,在出現(xiàn)的對(duì)話框中選擇“View/CovarianceAnalysis/correlation”,點(diǎn)擊ok,得到相關(guān)系數(shù)矩陣。計(jì)算各個(gè)解釋變量的相關(guān)系數(shù)，得到相關(guān)系數(shù)矩陣。表4.1相關(guān)系數(shù)矩陣變量X1X2X3X4X11.0000000.8097770.8720930.659239X20.8097771.0000000.7583220.641086X30.8720930.7583221.0000000.716374X40.6592390.6410860.7163741.000000由相關(guān)系數(shù)矩陣可以看出，解釋變量X2、X3之間存在較高的相關(guān)系數(shù)，證實(shí)確實(shí)存在嚴(yán)重的多重共線性。（2）多重共線性修正采用逐步回歸的辦法，檢驗(yàn)和回歸多重共線性問題。分別作Y對(duì)X1、X2、X3、X4的一元回歸，在命令窗口分別輸入LSYCX1，LSYCX2,LSYCX3，LSYCX4，并保存，整理結(jié)果如表4.2所示。表4.2一元回歸結(jié)果變量X1X2X3X4參數(shù)估計(jì)值1.9782240.3151200.31694612.54525t統(tǒng)計(jì)量8.63511112.474956.9224794.005547R20.7199830.8429240.6229880.3561910.7103270.8375080.6099880.333991其中，X2的方程最大，以X2為基礎(chǔ)，順次加入其它變量逐步回歸。在命令窗口中依次輸入：LSYCX2X1，LSYCX2X3,LSYCX2X4,并保存結(jié)果，整理結(jié)果如表4.3所示。表4.3加入新變量的回歸結(jié)果（一）變量變量X1X2X3X4X2，X10.711446(2.679575)0.230304(5.891959)0.866053X2，X30.258113(7.016265)0.087950(2.043471)0.853546X2，X40.312045(9.319239)0.293708（0.143226）0.831828經(jīng)比較，新加入X1的方程=0.866053，改進(jìn)最大，而且各個(gè)參數(shù)的t檢驗(yàn)顯著，選擇保留X1，再加入其它新變量逐步回歸，在命令框中依次輸入：LSYCX2X1X3，LSYCX2X1X4，保存結(jié)果，整理結(jié)果如表4.4所示。表4.4加入新變量的回歸結(jié)果（二）變量變量X1X2X3X4X2，X1，X30.603269(1.652919)0.227087(5.630196)0.024860(0.439370)0.862078X2，X1，X40.773017(2.741794)0.237243(5.833838)-1.364110(-0.701920)0.863581當(dāng)加入X3或X4時(shí)，均沒有所增加，且其參數(shù)是t檢驗(yàn)不顯著。從相關(guān)系數(shù)可以看出X3、X4與X1、X2之間相關(guān)系數(shù)較高，這說明X3、X4引起了多重共線性，予以剔除。當(dāng)取α=0.05時(shí)，tα/2(n-k-1)=2.048，X1、X2的系數(shù)t檢驗(yàn)均顯著，這是最后消除多重共線性的結(jié)果。修正多重共線性影響后的模型為=0.711446X+0.230304X（0.265507）（0.039088）t=(2.679575)(5.891959)R2=0.874983=0.866053在確定模型以后，進(jìn)行參數(shù)估計(jì)表4.5消除多重共線性后的回歸結(jié)果DependentVariable:YMethod:LeastSquaresDate:11/14/13Sample:131Includedobservations:31CoefficientStd.Errort-StatisticProb.

C-4316.82412795.42-0.3373730.7384X10.7114460.2655072.6795750.0122X20.2303040.0390885.8919590.0000R-squared0.874983

Meandependentvar114619.2AdjustedR-squared0.866053

S.D.dependentvar112728.1S.E.ofregression41257.10

Akaikeinfocriterion24.18480Sumsquaredresid4.77E+10

Schwarzcriterion24.32357Loglikelihood-371.8644

Hannan-Quinncriter.24.23004F-statistic97.98460

Durbin-Watsonstat1.893654Prob(F-statistic)0.000000五、異方差檢驗(yàn)在實(shí)際的經(jīng)濟(jì)問題中經(jīng)常會(huì)出現(xiàn)異方差這種現(xiàn)象，因此建立模型時(shí)，必須要注意異方差的檢驗(yàn)，否則，在實(shí)際中會(huì)失去意義。檢驗(yàn)異方差由表4.5的結(jié)果，按路徑“View/ResidualTests/HeteroskedasticityTests”，在出現(xiàn)的對(duì)話框中選擇Specification:White，點(diǎn)擊ok.得到White檢驗(yàn)結(jié)果如下。表5.1White檢驗(yàn)結(jié)果HeteroskedasticityTest:WhiteF-statistic3.676733

Prob.F(5,25)0.0125Obs*R-squared13.13613

Prob.Chi-Square(5)0.0221ScaledexplainedSS15.97891

Prob.Chi-Square(5)0.0069TestEquation:DependentVariable:RESID^2Method:LeastSquaresDate:11/14/13Time:21:48Sample:131Includedobservations:31CoefficientStd.Errort-StatisticProb.

C-1.10E+091.11E+09-0.9927790.3303X1-12789.3630151.30-0.4241730.6751X1^20.4207160.2943321.4293930.1653X1*X2-0.1018140.083576-1.2182160.2345X214604.525047.7012.8933010.0078X2^2-0.0024890.008030-0.3099720.7592R-squared0.423746

Meandependentvar1.54E+09AdjustedR-squared0.308495

S.D.dependentvar2.70E+09S.E.ofregression2.24E+09

Akaikeinfocriterion46.07313Sumsquaredresid1.26E+20

Schwarzcriterion46.35068Loglikelihood-708.1335

Hannan-Quinncriter.46.16360F-statistic3.676733

Durbin-Watsonstat1.542170Prob(F-statistic)0.012464從上表可以看出，nR=13.13613，由White檢驗(yàn)可知，在α=0.05下，查分布表，得臨界值χ(5)=11.0705,比較計(jì)算的統(tǒng)計(jì)量與臨界值，因?yàn)閚R=13.13613>χ(5)=11.0705,所以拒絕原假設(shè)，表明模型存在異方差。（2）異方差的修正①用WLS估計(jì)：選擇權(quán)重w=1/e1^2,其中e1=resid。在命令窗口中輸入genre1=resid,點(diǎn)回車鍵。在消除多重共線性后的回歸結(jié)果（表4.5的回歸結(jié)果）對(duì)話框中點(diǎn)擊Estimate/Options/WeithtedLS/TSLS,并在Weight中輸入1/e1^2,點(diǎn)確定，得到如下回歸結(jié)果。表5.2用權(quán)數(shù)1/e1^2的回歸結(jié)果DependentVariable:YMethod:LeastSquaresDate:11/14/13Time:21:49Sample:131Includedobservations:31Weightingseries:1/E1^2CoefficientStd.Errort-StatisticProb.

C-7074.873389.4944-18.164250.0000X10.7882770.01369257.570990.0000X20.2358060.000968243.67860.0000WeightedStatisticsR-squared0.999848

Meandependentvar31056.56AdjustedR-squared0.999837

S.D.dependentvar171821.4S.E.ofregression4.259384

Akaikeinfocriterion5.827892Sumsquaredresid507.9857

Schwarzcriterion5.966665Loglikelihood-87.33232

Hannan-Quinncriter.5.873128F-statistic92014.78

Durbin-Watsonstat1.663366Prob(F-statistic)0.000000UnweightedStatisticsR-squared0.871469

Meandependentvar114619.2AdjustedR-squared0.862288

S.D.dependentvar112728.1S.E.ofregression41832.86

Sumsquaredresid4.90E+10Durbin-Watsonstat1.853343②修正后的White檢驗(yàn)為在表5.2的回歸結(jié)果中，按路徑“View/ResidualTests/HeteroskedasticityTests”，在出現(xiàn)的對(duì)話框中選擇Specification:White，點(diǎn)擊ok.得到White檢驗(yàn)結(jié)果如下。表5.3修正后的White檢驗(yàn)結(jié)果HeteroskedasticityTest:WhiteF-statistic0.210748

Prob.F(2,28)0.8113Obs*R-squared0.459736

Prob.Chi-Square(2)0.7946ScaledexplainedSS0.595955

Prob.Chi-Square(2)0.7423TestEquation:DependentVariable:WGT_RESID^2Method:LeastSquaresDate:11/15/13Time:20:29Sample:131Includedobservations:31CollineartestregressorsdroppedfromspecificationCoefficientStd.Errort-StatisticProb.

C17.639915.9225942.9784100.0059WGT-256.0052728.8280-0.3512560.7280WGT^28.26192623.571550.3505040.7286R-squared0.014830

Meandependentvar16.38664AdjustedR-squared-0.055539

S.D.dependentvar29.69485S.E.ofregression30.50832

Akaikeinfocriterion9.765641Sumsquaredresid26061.21

Schwarzcriterion9.904414Loglikelihood-148.3674

Hannan-Quinncriter.9.810878F-statistic0.210748

Durbin-Watsonstat2.081320Prob(F-statistic)0.811251從上表可知nR==0.459736<χ(5)=11.0705，證明模型中的異方差已經(jīng)被消除了。異方差修正后的模型為=-7074.873+0.788277X1*+0.235806X2*389.49440.0136920.000968t=(-18.16425)(57.57099)(243.6786)R2=0.999848=0.999837F=92014.78其中X1*=1/e1^2*X1,X2*=1/e1^2*X2,e1=resid。六、自相關(guān)檢驗(yàn)與修正（1）DW檢驗(yàn)在顯著性水平α=0.05，查DW表，當(dāng)n=31，k=2時(shí)，得上臨界值d=1.27，下臨界值d=1.15，DW=1.663365。因?yàn)閐<DW<4-d，所以模型不存在序列自相關(guān)。由圖示法也可以看出隨機(jī)誤差項(xiàng)μi不存在自相關(guān)。下圖是殘差及一階滯后殘差相關(guān)圖。圖6.1殘差與其滯后一階殘差圖LM檢驗(yàn) 在表5.2的回歸結(jié)果中，按路徑“View/ResidualTests/SerialCorrelationLMTests”，在出現(xiàn)的對(duì)話框中選擇Lagstoinclude:1，點(diǎn)擊ok.得到LM檢驗(yàn)結(jié)果如下。表6.1LM檢驗(yàn)結(jié)果Breusch-GodfreySerialCorrelationLMTest:F-statistic0.809839

Prob.F(1,27)0.3761Obs*R-squared0.902738

Prob.Chi-Square(1)0.3420TestEquation:DependentVariable:RESIDMethod:LeastSquaresDate:11/14/13Time:21:50Sample:131Includedobservations:31Presamplemissingvaluelaggedresidualssettozero.Weightseries:1/E1^2CoefficientStd.Errort-StatisticProb.

C-766.3965937.0314-0.8178980.4206X10.0209900.0270700.7753900.4448X2-0.0012730.001716-0.7420020.4645RESID(-1)-0.0070920.007881-0.8999100.3761WeightedStatisticsR-squared0.029121

Meandependentvar-0.564513AdjustedR-squared-0.078755

S.D.dependentvar4.074747S.E.ofregression4.273921

Akaikeinfocriterion5.862855Sumsquaredresid493.1929

Schwarzcriterion6.047885Loglikelihood-86.87425

Hannan-Quinncriter.5.923170F-statistic0.269946

Durbin-Watsonstat1.683210Prob(F-statistic)0.846488UnweightedStatisticsR-squared-0.014569

Meandependentvar-4021.722AdjustedR-squared-0.127299

S.D.dependentvar40207.07S.E.ofregression42689.59

Sumsquaredresid4.92E+10Durbin-Watsonstat1.69E-08從上表可以看出，nR=0.902738，由LM檢驗(yàn)可知，在α=0.05下，查分布表，得臨界值χ(5)=11.0705,比較計(jì)算的統(tǒng)計(jì)量與臨界值，因?yàn)閚R=0.902738<χ(5)=11.0705,所以接受原假設(shè)，表明模型不存在自相關(guān)。

七、模型檢驗(yàn)經(jīng)濟(jì)意義檢驗(yàn)?zāi)Ｐ凸烙?jì)結(jié)果表明，在假定其他變量不變的情況下，當(dāng)景區(qū)固定資產(chǎn)每增長(zhǎng)1元時(shí)，旅游收入增加0.788277元；在假定其他變量不變的情況下，當(dāng)景區(qū)從業(yè)人員每增加1人時(shí)，旅游收入增加0.235806萬(wàn)元。這與理論分析判斷相一致。統(tǒng)計(jì)檢驗(yàn)（1）擬合優(yōu)度：由表中數(shù)據(jù)可得：R2=0.999848，修正的可決系數(shù)為=0.999837，這說明模型對(duì)樣本的擬合很好。（2）F檢驗(yàn)：針對(duì)H0：β1=β2=0，給定顯著性水平α=0.05，在F分布表中查出自由度為k=2和n-k-1=28的臨界值Fα(2，28)=3.34。由表中得到F=92014.78，由于F=92014.78>Fα(2，28)=3.34，應(yīng)拒絕原假設(shè)，說明回歸方程顯著，即“旅游景區(qū)固定資產(chǎn)”、“旅游從業(yè)人員”等變量聯(lián)合起來確實(shí)對(duì)“旅游景區(qū)營(yíng)業(yè)收入”有顯著影響。（3）t檢驗(yàn)：分別對(duì)H0：βj=0(j=1,2),給定顯著性水平α=0.05，查t分布表得自由度為n-k-1=28臨界值tα/2(n-k-1)=2.048。由表中數(shù)據(jù)可得，、對(duì)應(yīng)的t統(tǒng)計(jì)量分別為57.57099、243.6786，其絕對(duì)值均大于tα/2(n-k-1)=2.048，這說明應(yīng)該分別拒絕H0：βj=0(j=1,2),也就是說，當(dāng)在其他解釋變量不變的情況下，解釋變量“旅游景區(qū)固定資產(chǎn)”（X1）、“旅游從業(yè)人數(shù)”（X2）分別對(duì)被解釋變量“旅游景區(qū)營(yíng)業(yè)收入”(Y)影響顯著。八、附錄以下是多重共線性參數(shù)估計(jì)備表1對(duì)X回歸分析DependentVariable:YMethod:LeastSquaresDate:11/14/13Time:21:14Sample:131Includedobservations:31CoefficientStd.Errort-StatisticProb.

C-15595.6118604.86-0.8382550.4087X11.9782240.2290918.6351110.0000R-squared0.719983

Meandependentvar114619.2AdjustedR-squared0.710327

S.D.dependentvar112728.1S.E.ofregression60671.69

Akaikeinfocriterion24.92668Sumsquaredresid1.07E+11

Schwarzcriterion25.01920Loglikelihood-384.3636

Hannan-Quinncriter.24.95684F-statistic74.56515

Durbin-Watsonstat2.090544Prob(F-statistic)0.000000備表2對(duì)X回歸分析DependentVariable:YMethod:LeastSquaresDate:11/14/13Time:21:15Sample:131Includedobservations:31CoefficientStd.Errort-StatisticProb.

C15958.7311364.711.4042360.1709X20.3151200.02526012.474950.0000R-squared0.842924

Meandependentvar114619.2AdjustedR-squared0.837508

S.D.dependentvar112728.1S.E.ofregression45441.05

Akaikeinfocriterion24.34856Sumsquaredresid5.99E+10

Schwarzcriterion24.44108Loglikelihood-375.4027

Hannan-Quinncriter.24.37872F-statistic155.6243

Durbin-Watsonstat1.665119Prob(F-statistic)0.000000備表3對(duì)X回歸分析DependentVariable:YMethod:LeastSquaresDate:11/14/13Time:21:15Sample:131Includedobservations:31CoefficientStd.Errort-StatisticProb.

C53599.9515413.413.4774880.0016X30.3169460.0457856.9224790.0000R-squared0.622988

Meandependentvar114619.2AdjustedR-squared0.609988

S.D.dependentvar112728.1S.E.ofregression70399.77

Akaikeinfocriterion25.22411Sumsquaredresid1.44E+11

Schwarzcriterion25.31662Loglikelihood-388.9737

Hannan-Quinncriter.25.25427F-statistic47.92072

Durbin-Watsonstat1.724195Prob(F-statistic)0.000000備表4對(duì)X回歸分析DependentVariable:YMethod:LeastSquaresDate:11/14/13Time:21:15Sample:131Includedobservations:31CoefficientStd.Errort-StatisticProb.

C-143904.966622.99-2.1599890.0392X412.545253.1319704.0055470.0004R-squared0.356191

Meandependentvar114619.2AdjustedR-squared0.333991

S.D.dependentvar112728.1S.E.ofregression91996.75

Akaikeinfocriterion25.75923Sumsquaredresid2.45E+11

Schwarzcriterion25.85175Loglikelihood-397.2681

Hannan-Quinncriter.25.78939F-statistic16.04440

Durbin-Watsonstat1.829839Prob(F-statistic)0.000394備表5對(duì)X、X回歸分析DependentVariable:YMethod:LeastSquaresDate:11/14/13Time:21:15Sample:131Includedobservations:31CoefficientStd.Errort-StatisticProb.

C-4316.82412795.42-0.3373730.7384X20.2303040.0390885.8919590.0000X10.7114460.2655072.6795750.0122R-squared0.874983

Meandependentvar114619.2AdjustedR-squared0.866053

S.D.dependentvar112728.1S.E.ofregression41257.10

Akaikeinfocriterion24.18480Sumsquaredresid4.77E+10

Schwarzcriterion24.32357Loglikelihood-371.8644

Hannan-Quinncriter.24.23004F-statistic97.98460

Durbin-Watsonstat1.893654Prob(F-statistic)0.000000備表6對(duì)X、X回歸分析DependentVariable:YMethod:LeastSquaresDate:11/14/13Time:21:15Sample:131Includedobservations:31CoefficientStd.Errort-StatisticProb.

C16874.5310798.591.5626600.1294X20.2581130.0367887.0162650.0000X30.0879500.0430402.0434710.0505R-squared0.863310

Meandependentvar114619.2AdjustedR-squared0.853546

S.D.dependentvar112728.1S.E.ofregression43140.27

Akaikeinfocriterion24.27407Sumsquaredresid5.21E+10

Schwarzcriterion24.41284Loglikelihood-373.2480

Hannan-Quinncriter.24.31930F-statistic88.42123

Durbin-Watsonstat1.600090Prob(F-statistic)0.000000備表7對(duì)X、X回歸分析DependentVariable:YMethod:LeastSquaresDate:11/14/13Time:21:15Sample:131Includedobservations:31CoefficientStd.Errort-StatisticProb.

C10868.7937371.230.2908330.7733X20.3120450.0334849.3192390.0000X40.2937082.0506600.1432260.8871R-squared0.843039

Meandependentvar114619.2AdjustedR-squared0.831828