Prediction of Outstanding Claims: A Hierarchical Credibility Approach

Abstract

O. Prediction of outstanding claims: A hierarchical credibility approach. Scand. Actuarial J. 1991; 1: 25-47.We consider the problem of predicting unpaid losses in respect of incurred, but not yet reported claims in the case where the distribution of delays until notification may depend on the type of claim. A hierarchical credibility model is presented, where the distribution of delays corresponding to different types of claims, as well as other characteristics affecting the claims experience, are viewed as realizations of unobservable random variables. The proposed method is illustrated by a numerical example.

Robust Bayesian analysis of loss reserving data using scale mixtures distributions

“…, n. In the above log-ANOVA, log-ANCOVA and log-SS models, claim counts and inflation rates can be incorporated into the models. See Hesselager [13], Neuhaus [21] and Haastrup and Arjas [11] and Ntzoufras and Dellaportas [22] for details.…”

Quantification of automobile insurance liability: a Bayesian failure time approach

“…In this way we are essentially modelling "payments per claim finalized" (PPCF); see Taylor and Ashe (1983). Modelling of both claim counts and amounts has been advocated in the past; see Norberg (1986), Hesselager (1991), Neuhaus (1992) and Haastrup and Arjas (1996) for application in individual claim data, Taylor and Ashe (1983) for aggregated data and Ntzoufras and Dellaportas (2002) for a recent Bayesian two-stage hierarchical model.…”

Bayesian Modelling of Outstanding Liabilities Incorporating Claim Count Uncertainty

“…This has the advantage that an increase in accidents, which is expressed via an increase of claim counts in Table 2, results in an increase of total claims in Table 1. Modelling of both claim counts and amounts has been advocated in the past; see Norberg (1986), Hesselager (1991), Neuhaus (1992), and Haastrup and Arjas (1996) for application in individual claim data and Taylor and Ashe (1983) for aggregated data.…”