Robust Bayesian analysis of loss reserving data using scale mixtures distributions

It is vital for insurance companies to have appropriate levels of loss reserving to pay outstanding claims and related settlement costs. With many uncertainties and time lags inherently involved in the claims settlement process, loss reserving therefore must be based on estimates. Existing models and methods cannot cope with irregular and extreme claims and hence do not offer an accurate prediction of loss reserving. This paper extends the conventional normal error distribution in loss reserving modeling to a range of heavy-tailed distributions which are expressed by certain scale mixtures forms. This extension enables robust analysis and, in addition, allows an efficient implementation of Bayesian analysis via Markov chain Monte Carlo simulations. Various models for the mean of the sampling distributions, including the log-Analysis of Variance (ANOVA), log-Analysis of Covariance (ANCOVA) and state space models, are considered and the straightforward implementation of scale mixtures distributions is demonstrated using OpenBUGS.

[1]  R. Verrall.  Claims reserving and generalised additive models , 1996 .

[2]  Michael Merz,et al.  Paid-Incurred Chain Claims Reserving Method , 2010 .

[3]  S. T. Boris Choy,et al.  Scale Mixtures Distributions in Insurance Applications , 2003, ASTIN Bulletin.

[4]  S. T. Boris Choy,et al.  SCALE MIXTURES DISTRIBUTIONS IN STATISTICAL MODELLING , 2008 .

[5]  J. C. Hickman,et al.  “Principal Applications of Bayesian Methods in Actuarial Science: A Perspective”, Udi E. Makov, October 2001 , 2001 .

[6]  Adrian F. M. Smith,et al.  Exact and Approximate Posterior Moments for a Normal Location Parameter , 1992 .

[7]  Richard Verrall,et al.  Credibility Theory and Generalized Linear Models , 1997, ASTIN Bulletin.

[8]  Steven Haberman,et al.  Generalized linear models and actuarial science , 1996 .

[9]  Bradley P. Carlin,et al.  Bayesian measures of model complexity and fit , 2002 .

[10]  Edward W. Frees,et al.  Dependent Loss Reserving using Copulas , 2011 .

[11]  J. Beirlant,et al.  Lognormal Mixed Models for Reported Claims Reserves , 2006 .

[12]  R. Verrall,et al.  Chain ladder and maximum likelihood , 1991 .

[13]  O. Hesselager.  Prediction of Outstanding Claims: A Hierarchical Credibility Approach , 1991 .

[14]  Wai-Yin Wan,et al.  Bayesian Student-T Stochastic Volatility Models Via Scale Mixtures , 2009 .

[15]  A note on bounded influence in Bayesian analysis , 1995 .

[16]  Udi E. Makov,et al.  Principal Applications of Bayesian Methods in Actuarial Science , 2001 .

[17]  Leonard A. Stefanski.  A normal scale mixture representation of the logistic distribution , 1991 .

[18]  M. West.  Outlier Models and Prior Distributions in Bayesian Linear Regression , 1984 .

[19]  Greg Taylor.  Claims reserving in non-life insurance , 1985 .

[20]  S. T. Boris Choy,et al.  On Robust Analysis of a Normal Location Parameter , 1997 .

[21]  R. Verrall.  A Method for Modelling Varying Run-Off Evolutions in Claims Reserving , 1994 .

[22]  Piet de Jong,et al.  Forecasting Runoff Triangles , 2006 .

[23]  Claims Reserving Using Tweedie's Compound Poisson Model , 2003, ASTIN Bulletin.

[24]  Radford M. Neal.  Slice Sampling , 2003, The Annals of Statistics.

[25]  David P. M. Scollnik,et al.  Actuarial Modeling with MCMC and BUGs , 2001 .

[26]  R. Verrall.  A STATE SPACE REPRESENTATION OF THE CHAIN LADDER LINEAR MODEL , 1989 .

[27]  Richard Verrall,et al.  A Stochastic Model Underlying the Chain-Ladder Technique , 1998, British Actuarial Journal.

[28]  A. F. M. Smith,et al.  Bayesian methods in actuarial science , 1996 .

[29]  E. Arjas,et al.  Claims Reserving in Continuous Time; A Nonparametric Bayesian Approach , 1996, ASTIN Bulletin.

[30]  Ben Zehnwirth,et al.  Claims reserving, state-space models and the Kalman filter , 1983 .

[31]  Richard Verrall,et al.  Predictive Distributions of Outstanding Liabilities in General Insurance , 2006, Annals of Actuarial Science.

[32]  Enrique de Alba,et al.  Bayesian Estimation of Outstanding Claim Reserves , 2002 .

[33]  Alice X.D. Dong,et al.  Bayesian analysis of loss reserving using dynamic models with generalized beta distribution , 2013 .

[34]  Petros Dellaportas,et al.  Bayesian Modelling of Outstanding Liabilities Incorporating Claim Count Uncertainty , 2002 .