我國(guó)房地產(chǎn)行業(yè)的生產(chǎn)函數(shù)模型(編輯修改稿)
2025-02-14 00:09 本頁(yè)面
【文章內(nèi)容簡(jiǎn)介】 明該模型具有多重共線性。（2）異方差檢驗(yàn)下面我們首先通過(guò)ARCH檢驗(yàn)看我們可以看我們的模型是否存在異方差。ARCH Test:1階Fstatistic ProbabilityObs*Rsquared ProbabilityTest Equation:Dependent Variable: RESID^2Method: Least SquaresDate: 06/13/05 Time: 19:32Sample(adjusted): 1992 2003Included observations: 12 after adjusting endpointsVariableCoefficientStd. ErrortStatisticProb. CRESID^2(1)Rsquared Mean dependent varAdjusted Rsquared . dependent var. of regression Akaike info criterionSum squared resid Schwarz criterionLog likelihood FstatisticDurbinWatson stat Prob(Fstatistic)ARCH Test:2階Fstatistic ProbabilityObs*Rsquared ProbabilityTest Equation:Dependent Variable: RESID^2Method: Least SquaresDate: 06/13/05 Time: 19:30Sample(adjusted): 1993 2003Included observations: 11 after adjusting endpointsVariableCoefficientStd. ErrortStatisticProb. CRESID^2(1)RESID^2(2)Rsquared Mean dependent varAdjusted Rsquared . dependent var. of regression Akaike info criterionSum squared resid Schwarz criterionLog likelihood FstatisticDurbinWatson stat Prob(Fstatistic)ARCH Test:3階Fstatistic ProbabilityObs*Rsquared ProbabilityTest Equation:Dependent Variable: RESID^2Method: Least SquaresDate: 06/13/05 Time: 19:32Sample(adjusted): 1994 2003Included observations: 10 after adjusting endpointsVariableCoefficientStd. ErrortStatisticProb. CRESID^2(1)RESID^2(2)RESID^2(3)Rsquared Mean dependent varAdjusted Rsquared . dependent var. of regression Akaike info criterionSum squared resid Schwarz criterionLog likelihood FstatisticDurbinWatson stat Prob(Fstatistic)通過(guò)ARCH的3階的檢驗(yàn)，我們發(fā)現(xiàn)P值均大于5%，且t值檢驗(yàn)均小于2，則認(rèn)為不存在異方差。又由于是小樣本，所以不進(jìn)行WHITE檢驗(yàn)。（3）自相關(guān)檢驗(yàn)。DurbinWatson stat=，當(dāng)K’=2，n=13時(shí)，查表得du=，4du=，則模型不存在自相關(guān)。模型的修正對(duì)于模型具有的多重共線性，從經(jīng)濟(jì)意義上解釋，是由于勞動(dòng)力的增長(zhǎng)同資本的增長(zhǎng)隨時(shí)間的變化呈高度相關(guān)。假設(shè)已知規(guī)模報(bào)酬不變，在生產(chǎn)函數(shù)Q=f(L,K)=ALaKb中，即a+b=1。則原函數(shù)可以變?yōu)椋篞=ALaK1a，從而Q/k=A（L/K）a，其中Q/K為資本產(chǎn)出率，L/K為勞動(dòng)對(duì)資本的投入率。兩邊取對(duì)數(shù)得：Ln（Q/K）=LnA+aLn（L/K），這時(shí)求參數(shù)的估計(jì)值就避免了原模型的多重共線性。經(jīng)過(guò)調(diào)整得數(shù)據(jù)：表2年度房地產(chǎn)經(jīng)營(yíng)總收入Q單位：億元年度投資完成額K單位：億元平均從業(yè)人數(shù)單位：萬(wàn)人Ln（Q/K）Ln（L/K）1991199219931994199519961997199819992000200120022003一、數(shù)據(jù)平穩(wěn)性檢驗(yàn)（1）對(duì)于Ln（Q/K），0階差分，有趨勢(shì)和截距項(xiàng)，滯后3階：ADF Test Statistic 1% Critical Value* 5% Critical Value 10% Critical Value*MacKinnon critical values for rejection of hypothesis of a unit root.Augmented DickeyFuller Test EquationDependent Variable: D(LNQK)Method: Least SquaresDate: 06/14/05 Time: 11:28Sample(adjusted): 1995 2003Included observations: 9 after adjusting endpointsVariableCoefficientStd. ErrortStatisticProb.
公司管理相關(guān)推薦
鄂ICP備17016276號(hào)-1