計量經(jīng)濟學(xué)課程設(shè)計(編輯修改稿)
2025-02-09 18:06 本頁面
【文章內(nèi)容簡介】 業(yè)總產(chǎn)值的最優(yōu)擬合模型為：并設(shè)為原模型。表3 逐步回歸值值值值 序列相關(guān)性檢驗及改進由查表有故由檢驗知模型存在一階正自相關(guān)性，同時由殘差圖3，呈現(xiàn)有規(guī)律的波動也可認為原模型存在自相關(guān)性。圖3 殘差圖由偏相關(guān)系數(shù)檢驗結(jié)果（如圖4）知隨機干擾項只存在一階自相關(guān)。圖4 偏相關(guān)系數(shù)圖下面采用廣義差分法對模型進行處理：在軟件中直接使用廣義差分法估計自相關(guān)性，估計結(jié)果如圖5。Dependent Variable: LOG(Y)Method: Least SquaresDate: 06/02/12 Time: 15:52Sample(adjusted): 1991 2010Included observations: 20 after adjusting endpointsConvergence achieved after 8 iterationsVariableCoefficientStd. ErrortStatisticProb. CLOG(X2)LOG(X4)AR(1)Rsquared Mean dependent varAdjusted Rsquared . dependent var. of regression Akaike info criterionSum squared resid Schwarz criterionLog likelihood FstatisticDurbinWatson stat Prob(Fstatistic)Inverted AR Roots .79圖5 迭代估計法估計結(jié)果由圖5知根據(jù)檢驗知無法確定原模型是否仍然具有自相關(guān)性，因此再次用偏相關(guān)系數(shù)檢驗，由圖6的結(jié)果可得出原模型已經(jīng)不存在自相關(guān)性。圖6 相關(guān)系數(shù)和偏相關(guān)系數(shù)因此得出的回歸方程為：在上述結(jié)果的基礎(chǔ)上進行異方差性檢驗。首先，利用散點圖圖7對異方差性進行粗略檢驗：圖7 散點圖由圖7可認為原模型存在異方差性。再采用懷特檢驗，檢驗結(jié)果見圖8。White Heteroskedasticity Test:Fstatistic ProbabilityObs*Rsquared ProbabilityTest Equation:Dependent Variable: RESID^2Method: Least SquaresDate: 06/02/12 Time: 16:53Sample: 1991 2010Included observations: 20VariableCoefficientStd. ErrortStatisticProb. CLOG(X2)(LOG(X2))^2(LOG(X2))*(LOG(X4))LOG(X4)(LOG(X4))^2Rsquared Mean dependent varAdjusted Rsquared . dependent var. of regression Akaike info criterionSum squared resid Schwarz criterionLog likelihood FstatisticDurbinWatson stat Prob(Fstatistic)圖
醫(yī)療健康相關(guān)推薦
鄂ICP備17016276號-1