@techreport{eprints1803,
            type = {Working Paper},
       publisher = {Carlo Alberto Notebooks},
           title = {A Bayesian copula model for stochastic claims reserving},
          author = {Luca Regis},
     institution = {IMT Institute for Advanced Studies Lucca},
            year = {2011},
             url = {http://eprints.imtlucca.it/1803/},
        keywords = {stochastic claims reserving; bayesian copulas; solvency capital requirement; loss reserving; bayesian methods},
        abstract = {We present a full Bayesian model for assessing the reserve requirement of multiline Non-Life insurance companies. Bayesian models for claims reserving allow to account for expert knowledge in the evaluation of Outstanding Loss Liabilities, allowing the use of additional information at a low cost. This paper combines a standard Bayesian approach for the estimation of marginal distribution for the single Lines of Business for a Non-Life insurance company and a Bayesian copula procedure for the estimation of aggregate reserves. The model we present allows to "mix" own-assessments of dependence between LoBs at a company level and market-wide estimates provided by regulators. We illustrate results for the single lines of business and we compare standard copula aggregation for different copula choices and the Bayesian copula approach.}
}