On the (weak) dependence between risk proles in insurance data analysis.

• Autores: Francisco José Vázquez Polo, María del Carmen Martel Escobar, Agustín Hernández Bastida
• Localización: Anales de ASEPUMA, ISSN-e 2171-892X, Nº. 21, 2013, 13 págs.
• Idioma: inglés
• Enlaces
  • Texto completo (pdf)
• Resumen

  • A common assumption in the statistical model for Bayes premium in the insurance context, is the independence between risk profiles associated with random quantities considered. In this communication we consider the compound collective risk model in which the primary distribution is comprised of the Poisson-Lindley distribution with a  parameter, and where the secondary distribution is an exponential one with a  parameter. We consider the case of dependence between risk pro les (i.e., the parameters  and ), where the dependence is modelled by a Farlie-Gumbel-Morgenstern family. Statistical properties and some consequences on the Bayes premium of the structure dependence chosen are studied.