VAR appropriate lags=0

[Catherine85]

Dear all,

Please assist me in determining the VAR lag length.

I tried with 4, follow by 3, 2 and 1. The result I get is 0 (Based on AIC)

If this is the case, is there any alternatives to proceed/ what should I do on my dataset?

My dataset tested with ADF--> All stationary at 1st difference but non-stationary at level (at 5%)

Please help me. :(

Thanks.





VAR Lag Order Selection Criteria
Endogenous variables: DLNGDP DLNRCPT DLNREER DLNTRADE
Exogenous variables: C
Date: 08/28/10 Time: 15:01
Sample: 1980 2009
Included observations: 25

Lag LogL LR FPE AIC SC HQ

0 121.7899 NA* 9.50e-10* -9.423191* -9.228171* -9.369101*
1 130.0448 13.20778 1.80e-09 -8.803580 -7.828480 -8.533129
2 144.3289 18.28372 2.32e-09 -8.666313 -6.911132 -8.179500
3 155.4940 10.71849 4.68e-09 -8.279521 -5.744259 -7.576347
4 174.1810 11.95968 7.93e-09 -8.494480 -5.179138 -7.574945

* 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

[Szi]

Maybe forget about the lag length selection criteria (which are inherently problematic anyway), and choose the order of the VAR based on what you believe to be an appropriate length - i.e. justify it using theory rather than the statistical methods.

That seems quite commonplace