Premium
Bayesian solvency analysis with autocorrelated observations
Author(s) -
Mendoza M.,
E. NietoBarajas L.
Publication year - 2006
Publication title -
applied stochastic models in business and industry
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.413
H-Index - 40
eISSN - 1526-4025
pISSN - 1524-1904
DOI - 10.1002/asmb.626
Subject(s) - econometrics , solvency , quantile , bayesian probability , autocorrelation , economics , prior probability , computer science , statistics , finance , mathematics , artificial intelligence , market liquidity
Abstract Most financial institutions are required to comply with a minimum capital rule in order to face their obligations during a given period of time. Due to the random nature of the financial flows involved, the problem of assessing the amount of capital required must be analysed within a stochastic framework and the solution can be reduced to the estimation of a selected quantile. Given the financial impact of a specific capital requirement, a proper and careful choice of the underlying model is of great relevance. Here, we address the problem for insurance companies by proposing an autocorrelated model to describe the relative severity after a suitable transformation and compare the results with those of a model, which assumes independence among observations. We undertake a full Bayesian analysis and derive the reference priors for the models. Results are illustrated with a real data set from the Mexican insurance industry. Copyright © 2006 John Wiley & Sons, Ltd.