The aim of this study is to develop efficient estimation methods of the maximum likelihood estimation method for the q-Pareto distribution (q-PD) parameters that are based on numerical methods to maximize the log-likelihood of the q-PD. And the q-PD parameters is a q-exponential distribution, finding applications in many fields such as:  seismology (earthquake patterns).

This study compares different methods to predict ocean surface currents in a fast-changing strait. Using over three years of radar observations, we tested both traditional and modern data-driven approaches. The results show that advanced learning models give the most accurate forecasts, especially for several hours ahead. We also find that including enough past information improves predictions. These findings can help improve maritime safety and support better decision-making in coastal areas.