This study aimed to use panel dynamic models (Panel NARDL, Panel ARDL) in statistical estimation to measure the effects of time variation and cross-sectional data simultaneously. its application is to measure the impact of GDP, exchange rate and oil price on inflation rate for North African countries (Egypt, Libya, Tunisia, Algeria, Morocco) during the period from 1990 to 2022. Using Hsiao test was done for ensuring non-total homogeneity and non-homogeneity of parameters and constants of the panel model. By estimating both the Panel ARDL and Panel NARDL models for five countries, the results indicate that the Panel NARDL model is better than the Panel ARDL and it is more suitable for the data, it has the highest asymmetric error correction term (ECT (-1)) =57911.-, R2=0.601, and is lower in terms of the AIC=1.833 criterions. The empirical results clearly show also that only in the long run, the positive shocks of oil price and exchange rate affect inflation rate in the NARDL model. In the short run, there is no effect of exchange rate or oil price shocks in the NARDL model.