Análisis del Riesgo de Caída de Cartera en Seguros: Metodologías de “Inteligencia Artificial” vs “Modelos Lineales Generalizados”

Abstract

This paper describes the use of two different methodological approaches for risk assessment: nonparametric coming from Artificial Intelligence techniques and, in contrast, Generalized Linear Models from statistical parametric. Both practical applications will analyze Lapse Risk, one of the measurable risks that insurance sector must take into account according to Solvency II. Results and conclusions show a new approach and how these techniques can be used by insurance companies as an improvement in Risk Management; encouraging the insurance sector to investigate new methodologies and techniques to deal requirements demanded by Solvency II regulation.