VAR Lag Order Selection Criteria
Posted: Mon Nov 21, 2011 11:08 am
Hi everybody
I want to do a granger causality test between FDI and TFP and its components, I want to calculate the appropriate lags but I see that I can calculate VAR Lag Order Selection Criteria using two methods; Quick - Estimate VAR and on Proc- Make Vector Autoregression. I did both and the results are below ( in the order mention). Because the first on indicate to use one lag and the second one 5. Can someone please help me.
VAR Lag Order Selection Criteria
Endogenous variables: FDIGDP
Exogenous variables: TFP TECCH EFFCH
Date: 11/21/11 Time: 13:55
Sample: 1970 2000
Included observations: 25
Lag LogL LR FPE AIC SC HQ
0 -29.75800 NA 0.805675 2.620640 2.766905 2.661208
1 -13.44844 27.40005* 0.237111* 1.395875* 1.590895* 1.449966*
2 -13.04477 0.645878 0.249367 1.443581 1.687357 1.511194
3 -11.89999 1.740064 0.247504 1.431999 1.724529 1.513135
4 -11.56566 0.481431 0.262564 1.485253 1.826538 1.579911
5 -9.888556 2.280864 0.250700 1.431084 1.821125 1.539265
6 -9.796415 0.117939 0.272425 1.503713 1.942509 1.625416
* 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
VAR Lag Order Selection Criteria
Endogenous variables: FDIGDP
Exogenous variables: C EFFCH TFP TECCH
Date: 11/21/11 Time: 14:00
Sample: 1970 2000
Included observations: 25
Lag LogL LR FPE AIC SC HQ
0 -29.51506 NA 0.857358 2.681205 2.876225 2.735295
1 -13.10939 26.24907* 0.250659* 1.448751 1.692526* 1.516364*
2 -12.74081 0.560234 0.264725 1.499265 1.791795 1.580400
3 -11.57047 1.685294 0.262665 1.485637 1.826923 1.580295
4 -10.99138 0.787561 0.273823 1.519310 1.909351 1.627491
5 -9.007876 2.538883 0.255770 1.440630* 1.879425 1.562333
6 -8.823255 0.221545 0.276728 1.505860 1.993411 1.641086
* 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
I want to do a granger causality test between FDI and TFP and its components, I want to calculate the appropriate lags but I see that I can calculate VAR Lag Order Selection Criteria using two methods; Quick - Estimate VAR and on Proc- Make Vector Autoregression. I did both and the results are below ( in the order mention). Because the first on indicate to use one lag and the second one 5. Can someone please help me.
VAR Lag Order Selection Criteria
Endogenous variables: FDIGDP
Exogenous variables: TFP TECCH EFFCH
Date: 11/21/11 Time: 13:55
Sample: 1970 2000
Included observations: 25
Lag LogL LR FPE AIC SC HQ
0 -29.75800 NA 0.805675 2.620640 2.766905 2.661208
1 -13.44844 27.40005* 0.237111* 1.395875* 1.590895* 1.449966*
2 -13.04477 0.645878 0.249367 1.443581 1.687357 1.511194
3 -11.89999 1.740064 0.247504 1.431999 1.724529 1.513135
4 -11.56566 0.481431 0.262564 1.485253 1.826538 1.579911
5 -9.888556 2.280864 0.250700 1.431084 1.821125 1.539265
6 -9.796415 0.117939 0.272425 1.503713 1.942509 1.625416
* 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
VAR Lag Order Selection Criteria
Endogenous variables: FDIGDP
Exogenous variables: C EFFCH TFP TECCH
Date: 11/21/11 Time: 14:00
Sample: 1970 2000
Included observations: 25
Lag LogL LR FPE AIC SC HQ
0 -29.51506 NA 0.857358 2.681205 2.876225 2.735295
1 -13.10939 26.24907* 0.250659* 1.448751 1.692526* 1.516364*
2 -12.74081 0.560234 0.264725 1.499265 1.791795 1.580400
3 -11.57047 1.685294 0.262665 1.485637 1.826923 1.580295
4 -10.99138 0.787561 0.273823 1.519310 1.909351 1.627491
5 -9.007876 2.538883 0.255770 1.440630* 1.879425 1.562333
6 -8.823255 0.221545 0.276728 1.505860 1.993411 1.641086
* 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