单方程第6题误差修正模型ECM-例5.10财政支出和财政收入协整关系检验.docxVIP

财政收入和财政支出的关系图：新建workfile，注意开始日期和截止日期：把数据载入workfile中，取名：f_ex 和 f_in.把两个变量作为一个组打开，点击view/garph画图，直接点击ok。单位根检验表面：财政支出和财政收入都是一阶单整，可以探讨两者的协整关系。具体操作步骤：先单独打开f_ex,点击view/unit root test.然后点击ok。如下图所示。接下来，要再次重复这个动作，只是选项中药选择1 st difference。如下图所示点击ok。单位根检验如下表：Null Hypothesis: F_EX has a unit rootExogenous: ConstantLag Length: 11 (Automatic - based on SIC, maxlag=14)t-StatisticProb.*Augmented Dickey-Fuller test statistic17.491491.0000Test critical values:1% level-3.4624125% level-2 level-2.574309Null Hypothesis: D(F_EX) has a unit rootExogenous: Constant, Linear TrendLag Length: 13 (Automatic - based on SIC, maxlag=14)t-StatisticProb.*Augmented Dickey-Fuller test statistic-4.5954680.0014Test critical values:1% level-4.0043655% level-3 level-3.139924Null Hypothesis: F_IN has a unit rootExogenous: Constant, Linear TrendLag Length: 11 (Automatic - based on SIC, maxlag=14)t-StatisticProb.*Augmented Dickey-Fuller test statistic8.4344061.0000Test critical values:1% level-4.0036755% level-3 level-3.139728Null Hypothesis: D(F_IN) has a unit rootExogenous: Constant, Linear TrendLag Length: 12 (Automatic - based on SIC, maxlag=14)t-StatisticProb.*Augmented Dickey-Fuller test statistic-4.6826840.0010Test critical values:1% level-4.0041325% level-3 level-3.139858把f_ex和f_in以一个group打开，点击view/contegration test/Johansen system…..然后点击ok。两者之间的协整关系检验如下表：Data Trend:NoneNoneLinearLinearQuadraticTest TypeNo InterceptInterceptInterceptInterceptInterceptNo TrendNo TrendNo TrendTrendTrendTrace22211Max-Eig22211再次点击view/granger causality两者之间格兰杰因果关系检验如下表：Pairwise Granger Causality TestsDate: 01/04/14 Time: 19:37Sample: 1990M01 2007M12Lags: 2?Null Hypothesis:ObsF-StatisticProb.??F_EX does not Granger Cause F_IN?214?51.54936.E-19?F_IN does not Granger Cause F_EX?63.32413.E-22接下来要对f_ex和f_in进行差分。把这两个变量作为一个组打开，点击Default选项下的log，把数据复制到excel中，重新命名ln(f_ex)和ln(f_in),按照上面说的步骤把数据重新载入eviews，或者之前就在excel中对数据进行取对数，需要注意的是，上面的画图、协整关系、格兰杰因果关系检验也可以对ln(f_ex)和ln(f_in)进行。建立方程：点击proc/make