In order to characterize seismically the Coquimbo Region, Chile, data from 1943 to 2020 were considered. The region was divided into 30 zones of 0.5 • latitude and 0.5 • longitude and the non-extensive statistical physics was used. Both, the Sotolongo-Costa-Posadas and Mathai models were proposed to analyze the magnitude-frequency distribution. The subdivision into cells of the catalog allowed to demonstrate that systems with value of q ∼ 1 present exponential behavior, while with the superposition of subsystems it was expected to obtain q > 1, supporting the su-perstatistical model. Thus, by subdividing the Coquimbo region into the South and North regions, we find that in both the entropic index is greater than one, q > 1, however , according to the value obtained, in the southern sector the long-range effects are greater than in the northern sector. This means that both sectors ae well described by the nonextensive statistical model, either by SCP model or Mathai model; the entropic index is q > 1 and in both cases R 2 > 0.99. As the Coquimbo region is considered as a all, the nonextensive statistical distribution is the more adequated. With respect to the probabilistic seismic hazard assessment, we showed that the Mathai’s model has the better fit. Thus, in the frequency-interevent time distribution was used for different limite magnitudes values. Our analysis shows that the probability occurrence of an seismic event in the northern sector of the region is lower than in the southern sector for the same period; in the former the behavior is of Poissonian type. Thus, the region’s northern and southern zones show different behaviors in terms of earthquake occurrence probability.