Procesos puntuales espaciales y sus aplicaciones : procesos puntuales de Thomas.

Abstract

Traffic accidents are the second leading cause of death in Ecuador accor ding to Primicias magazine. Until 2021, they were the eighth leading cause, which is quite surprising as in just three years, they moved up six positions. In this project, a study of point processes in space and on the network is conducted to model the intensity of traffic accidents recor ded in the Eugenio Espejo zonal administration between 2017 and 2023. Using records up to 2022, the intensity is estimated and, subsequently, new data is simulated based on this information. These simulated data are compared with the accidents recorded in 2023 to evaluate the accu racy of the model. As a result, an estimation of the sectors with the highest risk of traf f ic accidents in the area was obtained, patterns and concentrations of accidents, as well as clusters, were identified. The results of the spatial analysis are compared with those obtained from the network analysis, highlighting the strengths and limitations of each approach. Additionally, a Thomas process is simulated and modeled to analyze the occurrence of accidents and their clustering patterns. This stochastic process allows studying the clustering patterns and provides a deeper understanding of how and why accident clusters form.