haoyun010 发表于 2013-1-11 19:42 
协整检验的滞后期为无约束的VAR模型确定的最佳滞后期减去一。
谢谢
VAR Lag Order Selection Criteria
Endogenous variables: FDI IOV
Exogenous variables: C
Date: 01/11/13 Time: 20:20
Sample: 1990 2011
Included observations: 17
Lag LogL LR FPE AIC SC HQ
0 -336.4097 NA 6.69e+14 39.81291 39.91093 39.82265
1 -297.8863 63.45038* 1.16e+13 35.75133 36.04540* 35.78056
2 -295.3963 3.515235 1.43e+13 35.92898 36.41910 35.97770
3 -290.2926 6.004415 1.34e+13 35.79913 36.48530 35.86733
4 -283.1298 6.741393 1.06e+13* 35.42704* 36.30927 35.51473*
5 -281.4670 1.173774 1.80e+13 35.70200 36.78028 35.80918
* indicates lag order selected by the criterion
LR: sequential modified LR test statistic (each test at 5% level)
FPE: Final prediction error
AIC: Akaike information criterion
SC: Schwarz information criterion
HQ: Hannan-Quinn information criterion
这个的阶数应该为四吗?感觉用VAR Lag Order Selection Criteria,如果不同的最大值得到的最优滞后期都不相同,这个应该怎么办呢?