请教一下，进行格兰杰因果检验要用哪个最优滞后阶数？

6249

收藏 2018-04-27

VAR Lag Order Selection Criteria
Endogenous variables: PF PX
Exogenous variables: C
Date: 04/27/18 Time: 21:11
Sample: 1/04/2006 12/31/2008
Included observations: 554

Lag         LogL            LR             FPE                      AIC                   SC                HQ

0 -10254.94       NA          4.13e+13    37.02866    37.04424    37.03475
1 -8774.520    2944.805    2.00e+11    31.69863    31.74538*    31.71689*
2 -8767.487    13.93798    1.98e+11    31.68768    31.76561    31.71812
3 -8762.757    9.341309    1.98e+11    31.68504    31.79414    31.72766
4 -8753.080    19.03885*    1.94e+11    31.66455    31.80482    31.71935
5 -8748.350    9.271496    1.93e+11*    31.66191*    31.83335    31.72889
6 -8744.405    7.706006    1.93e+11    31.66211    31.86472    31.74126
7 -8742.120    4.446618    1.94e+11    31.66830    31.90208    31.75963
8 -8738.490    7.035985    1.95e+11    31.66964    31.93459    31.77315

* 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
请问进行格兰杰因果检验是的滞后期应该是4还是5