The minimum lag for the exogenous variable is determined through a contemporaneous structure, using a non-synchronism correction matrix (Table 5.1-2) and maximum lag is obtained from the maximum detected lag where the cross-correlation coefficient is found to be significant in s-statistics of the Cheung and Ng (1996) CCF causality test. The estimation is based on full sample data (3130 observations). The rejection of the causality hypothesis at 5% and 1% level of significance are represented by marks * and ** respectively. Mark “A” represents a case where the CCF causality test is not detected, by cross-correlation, to be significant in s-statistics, thus, models (21& 22) are not applied. Mark “B” represents a case where the convergence of estimated models is not achieved. Mark “-“ displays not applicable cases in Panel B and Panel D, where oil price cannot be adjusted on itself.