On Modelling Insurance Data by Using a Generalized Lognormal Distribution // Sobre la modelización de datos de seguros usando una distribución lognormal generalizada
View/ Open
Metadata
Show full item recordPalabras clave
Heavy-tailedinsurance
lognormal distribution
loss distribution
seguros
distribución lognormal
función de perdidas
colas pesadas
Publication date
2014Abstract
In this paper, a new heavy-tailed distribution is used to model data with a strong right tail, as often occurs in practical situations. The distribution proposed is derived from the lognormal distribution, by using the Marshall and Olkin procedure. Some basic properties of this new distribution are obtained and we present situations where this new distribution correctly reflects the sample behaviour for the right tail probability. An application of the model to dental insurance data is presented and analysed in depth. We conclude that the generalized lognormal distribution proposed is a distribution that should be taken into account among other possible distributions for insurance data in which the properties of a heavy-tailed distribution are present.------------------------------------Presentamos una nueva distribución lognormal con colas pesadas que se adapta bien a muchas situaciones prácticas en el campo de los seguros. Utilizamos el procedimiento de Marshall y Olkin para gene ...
In this paper, a new heavy-tailed distribution is used to model data with a strong right tail, as often occurs in practical situations. The distribution proposed is derived from the lognormal distribution, by using the Marshall and Olkin procedure. Some basic properties of this new distribution are obtained and we present situations where this new distribution correctly reflects the sample behaviour for the right tail probability. An application of the model to dental insurance data is presented and analysed in depth. We conclude that the generalized lognormal distribution proposed is a distribution that should be taken into account among other possible distributions for insurance data in which the properties of a heavy-tailed distribution are present.------------------------------------Presentamos una nueva distribución lognormal con colas pesadas que se adapta bien a muchas situaciones prácticas en el campo de los seguros. Utilizamos el procedimiento de Marshall y Olkin para generar tal distribución y estudiamos sus propiedades básicas. Se presenta una aplicación de la misma para datos de seguros dentales que es analizada en profundidad, concluyendo que tal distribución debería formar parte del catálogo de distribuciones a tener en cuenta para la modelización de datos en seguros cuando hay presencia de colas pesadas.