协整检验、格兰杰因果检验怎么做啊,

平稳性检验就是单位根检验
先来看一下序列X是否平稳
Null Hypothesis: X has a unit root
Exogenous: None
Lag Length: 0 (Automatic based on SIC, MAXLAG=2)

 t-Statistic Prob.*

Augmented Dickey-Fuller test statistic 9.533462 1.0000
Test critical values: 1% level-2.792154
 5% level-1.977738
 10% level-1.602074


原假设是存在单位根,序列是不平稳的.看是我们看ADF统计量值9.53,比10%水平下的值都要大,所以是接受原假设的,所以序列X是不平稳的.
再来看序列Y
 t-Statistic Prob.*

Augmented Dickey-Fuller test statistic3.826736 0.9990
Test critical values: 1% level-2.847250
 5% level-1.988198
 10% level-1.600140
同X一样,序列Y也是非平稳的.
协整检验就有点麻烦,先要对X和Y做差分,我这里是做了二阶差分才发现X,Y是平稳的,二阶差分后的序列定义为iix和iiy
对x和y序列做普通最小二乘回归
ls y c x
然后对残差序列做单位根检验
Null Hypothesis: E has a unit root
Exogenous: None
Lag Length: 0 (Automatic based on SIC, MAXLAG=2)

 t-Statistic Prob.*

Augmented Dickey-Fuller test statistic -1.236694 0.1853
Test critical values: 1% level-2.792154
 5% level-1.977738
 10% level-1.602074
可以看出,检验统计量-1.24大于10%水平下的-1.6,可以认为残差序列为非平稳序列,所以x和y不具有协整关系.
最后来看格兰杰因果检验
Pairwise Granger Causality Tests
Date: 06/06/12 Time: 09:20
Sample: 1999 2010
Lags: 2

Null Hypothesis:Obs F-StatisticProb.

Y does not Granger Cause X 10 0.608440.5801
X does not Granger Cause Y 0.831180.4879
可以看出F统计量是不显著的,X、Y不存在格兰杰成因.