Credibility Theory

This survey of actuarial credibility theory traces its origins, describes its evolutionary history, gives an account of its main issues and results, and assesses its merits with a view to related work in statistical science. Key-words: Experience rating, limited fluctuation theory, greatest accuracy theory, Bayes, linear Bayes, empirical Bayes, hierarchical models, Hilbert space methods, recursive computation, history of actuarial science. A. Credility what it was and what it is. In actuarial parlance the term credibility was originally attached to experience rating formulas that were convex combinations (weighted averages) of individual and class estimates of the individual risk premium. Credibility theory, thus, was the branch of insurance mathematics that explored model-based principles for construction of such formulas. The development of the theory brought it far beyond the original scope so that in today’s usage credibility covers more broadly linear estimation and prediction in latent variable models. B. The origins. The advent of credibility dates back to Whitney [46], who in 1918 addressed the problem of assessing the risk premium m, defined as the expected claims expenses per unit of risk exposed, for an individual risk selected from a portfolio (class) of similar risks. Advocating the combined use of individual risk experience and class risk experience, he proposed that the premium rate be a weighted average of the form m̄ = z m̂ + (1− z) μ , (1) where m̂ is the observed mean claim amount per unit of risk exposed for the individual contract and μ is the corresponding overall mean in the insurance portfolio. Whitney viewed the risk premium as a random variable. In the language of modern credibility theory, it is a function m(Θ) of a random element Θ representing the unobservable characteristics of the individual risk. The random nature of Θ expresses the notion of heterogeneity; the individual risk is a random selection from a portfolio of similar but not identical risks, and the distribution of Θ describes the variation of individual risk characteristics across the portfolio.

[1]  William S. Jewell,et al.  Regularity Conditions For Exact Credibility , 1975, ASTIN Bulletin.

[2]  Ragnar Norberg,et al.  Empirical Bayes credibility , 1980 .

[3]  W. Neuhaus.  Inference about parameters in empirical linear Bayes estimation problems , 1984 .

[4]  Ragnar Norberg,et al.  Hierarchical credibility: analysis of a random effect linear model with nested classification , 1986 .

[5]  P. A. V. B. Swamy,et al.  Statistical Inference in Random Coefficient Regression Models , 1971 .

[6]  T. Başar,et al.  A New Approach to Linear Filtering and Prediction Problems , 2001 .

[7]  P. Diaconis,et al.  Conjugate Priors for Exponential Families , 1979 .

[8]  H. Robbins.  An Empirical Bayes Approach to Statistics , 1956 .

[9]  Serge L. Wind.  AN EMPIRICAL BAYES APPROACH TO MULTIPLE LINEAR REGRESSION , 1973 .

[10]  B. Sundt.  Recursive credibility estimation , 1981 .

[11]  W. S. Jewell,et al.  The Use of Collateral Data in Credibility Theory: A Hierarchical Model , 1975 .

[12]  A. Bailey,et al.  AN INTRODUCTION TO CREDIBILITY THEORY , 2000 .

[13]  A. Bailey.  CREDIBILITY PROCEDURES LAPLACE'S GENERALIZATION OF BAYES' RULE AND THE COMBINATION OF COLLATERAL KNOWLEDGE WITH OBSERVED DATA , 2001 .

[14]  Thomas Witting.  The Linear Markov Property in Credibility Theory , 1987, ASTIN Bulletin.

[15]  Charles A. Hachemeister,et al.  CREDIBILITY FOR REGRESSION MODELS WITH APPLICATION TO TREND (REPRINT) , 2000 .

[16]  M. Mashayekhi.  On asymptotic optimality in empirical Bayes credibility , 2002 .

[17]  Ronald Bremer,et al.  Estimation of variance components and applications , 1988 .

[18]  Michael Goldstein,et al.  Approximate Bayes Solutions to Some Nonparametric Problems , 1975 .

[19]  B. Sundt.  A hierarchical credibility regression model , 1979 .

[20]  W. Jewell.  Two classes of covariance matrices giving simple linear forecasts , 1976 .

[21]  The mean credibility formula is a bayes rule , 1977 .

[22]  G. Taylor.  Experience Rating with Credibility Adjustment of the Manual Premium , 1974, ASTIN Bulletin: The Journal of the International Actuarial Association.

[23]  Hans Bühlmann,et al.  Experience rating and credibility , 1967, ASTIN Bulletin.

[24]  W. Neuhaus.  Choice of statistics in linear Bayes estimation , 1985 .

[25]  Rates of risk convergence of empirical linear Bayes estimators , 1992 .

[26]  Bjørn Sundt.  A multi-level hierarchical credibility regression model , 1980 .

[27]  T. Ferguson.  A Bayesian Analysis of Some Nonparametric Problems , 1973 .