【正文】
圖顯然我們發(fā)現(xiàn)這里取q=1或者2比較合適,即MA(1)和MA(2)過程。八、變量協(xié)整關(guān)系檢驗(yàn)為了檢驗(yàn)變量之間的協(xié)整關(guān)系,首先對變量進(jìn)行單位根檢驗(yàn),我們選取的方法是ADF檢驗(yàn),先以GDP為例,使用EVIEWS軟件操作得到結(jié)果如表5所示:表3:level條件下gdp單位根檢驗(yàn):Null Hypothesis: GDP has a unit rootExogenous: Constant, Linear TrendLag Length: 2 (Automatic based on SIC, MAXLAG=2)tStatisticTest critical values:1% level5% level10% level*MacKinnon (1996) onesided pvalues.Augmented DickeyFuller Test EquationDependent Variable: D(GDP)Method: Least SquaresDate: 01/06/10 Time: 16:48Sample (adjusted): 1955 2006Included observations: 52 after adjustmentsVariableCoefficientStd. ErrortStatisticProb.Prob.*Augmented DickeyFuller test statisticProb.*Augmented DickeyFuller test statistic 接受H0,不平穩(wěn)d_gdp (1 dif) 接受H0,不平穩(wěn)d_gdp (2 dif) 拒絕H 1,平穩(wěn)d_wj (level)首先,我們選用EG兩步法,對gdp(國內(nèi)生產(chǎn)總值)與wj(文教支出)進(jìn)行協(xié)整檢驗(yàn)。Test critical values:1% level5% level10% level*MacKinnon (1996) onesided pvalues.Augmented DickeyFuller Test EquationDependent Variable: D(RESID01)Method: Least SquaresDate: 01/06/10 Time: 21:42Sample (adjusted): 1962 2006Included observations: 45 after adjustments可見,拒絕零假設(shè)。國內(nèi)生產(chǎn)總值的提高必然會(huì)帶來稅收等增加,從而文教支出也會(huì)隨之增長。CCZZCRsquaredMean dependent varAdjusted Rsquared. dependent var. of regressionAkaike info criterionSum squared residSchwarz criterionLog likelihoodHannanQuinn criter.FstatisticDurbinWatson statProb(Fstatistic)Dependent Variable: WJMethod: Least SquaresDate: 01/06/10 Time: 21:49Sample: 1953 2006Included observations: 54VariableCoefficientStd. ErrortStatisticProb.Prob.*Augmented DickeyFuller test statistic也就是說財(cái)政支出與文教支出的增量間不存在長期均衡關(guān)系。CGDPCZZCRsquaredMean dependent varAdjusted Rsquared. dependent var. of regressionAkaike info criterionSum squared residSchwarz criterionLog likelihoodHannanQuinn criter.FstatisticDurbinWatson statProb(Fstatistic)Dependent Variable: WJMethod: Least SquaresDate: 01/06/10 Time: 21:56Sample: 1953 2006Included observations: 54VariableCoefficientStd. ErrortStatisticProb.Prob.*Augmented DickeyFuller test statistic也就是說國內(nèi)生產(chǎn)總值與文教支出的增量以及財(cái)政支出增量間存在長期均衡關(guān)系。Trace test indicates 2 cointegrating eqn(s) at the level* denotes rejection of the hypothesis at the level基于以上研究,我們進(jìn)一步做出