More importantly, the TYDL technique avoids the need for the preliminary tests for cointegration and is applicable irrespective of the integration or cointegration present in the system, because the singularity involved in the asymptotic distributions of the LS estimators is removed by fitting augmented VARL process whose order exceeds the true lag order by the highest degree of integration in the system.
In the TYDL methodology this implies that we have to augment the VARL by overfitting by additional two lag orders on top of the optimal order of lag for the VAR system.
Estimation of the augmented VARL (4) System and the Results of Hypotheses Tests