PARSIMONIOUS PARAMETERIZATION OF AGE-PERIOD-COHORT MODELS BY BAYESIAN SHRINKAGE

Abstract Age-period-cohort models used in life and general insurance can be over-parameterized, and actuaries have used several methods to avoid this, such as cubic splines. Regularization is a statistical approach for avoiding over-parameterization, and it can reduce estimation and predictive variances compared to MLE. In Markov Chain Monte Carlo (MCMC) estimation, regularization is accomplished by the use of mean-zero priors, and the degree of parsimony can be optimized by numerically efficient out-of-sample cross-validation. This provides a consistent framework for comparing a variety of regularized MCMC models, such as those built with cubic splines, linear splines (as ours is), and the limiting case of non-regularized estimation. We apply this to the multiple-trend model of Hunt and Blake (2014).

[1]  Steven Haberman,et al.  A cohort-based extension to the Lee-Carter model for mortality reduction factors , 2006 .

[2]  L. Gordon,et al.  The Gamma Function , 1994, Series and Products in the Development of Mathematics.

[3]  Gareth W. Peters,et al.  A unified approach to mortality modelling using state-space framework: characterisation, identification, estimation and forecasting , 2016, Annals of Actuarial Science.

[4]  G. C. Taylor.  Separation of Inflation and other Effects from the Distribution of Non-Life Insurance Claim Delays , 1977, ASTIN Bulletin.

[5]  Alan E. Gelfand,et al.  Model Determination using sampling-based methods , 1996 .

[6]  E. Carlson.  Who are the Lucky Few , 2008 .

[7]  Matthew P. Wand,et al.  Fully simplified multivariate normal updates in non-conjugate variational message passing , 2014, J. Mach. Learn. Res..

[8]  G. Venter,et al.  TRANSFORMED BETA AND GAMMA DISTRIBUTIONS AND AGGREGATE LOSSES , 1999 .

[9]  H. Keisler.  Elementary Calculus: An Infinitesimal Approach , 1976 .

[10]  Scott D. Brown,et al.  A simple introduction to Markov Chain Monte–Carlo sampling , 2016, Psychonomic bulletin & review.

[11]  B. Efron,et al.  Data Analysis Using Stein's Estimator and its Generalizations , 1975 .

[12]  Arthur E. Hoerl,et al.  Ridge Regression: Biased Estimation for Nonorthogonal Problems , 2000, Technometrics.

[13]  Stuart A. Klugman,et al.  Loss Models: From Data to Decisions, 2nd edition , 2004 .

[14]  D. Blake,et al.  A General Procedure for Constructing Mortality Models , 2014 .

[15]  Nicole Radde,et al.  Hamiltonian Monte Carlo methods for efficient parameter estimation in steady state dynamical systems , 2014, BMC Bioinformatics.

[16]  Aki Vehtari,et al.  Practical Bayesian model evaluation using leave-one-out cross-validation and WAIC , 2015, Statistics and Computing.

[17]  Katrien Antonio,et al.  Bayesian Poisson log-bilinear models for mortality projections with multiple populations , 2015 .

[18]  Jianming Ye.  On Measuring and Correcting the Effects of Data Mining and Model Selection , 1998 .

[19]  Ronald Lee,et al.  Modeling and forecasting U. S. mortality , 1992 .

[20]  David Blake,et al.  A Quantitative Comparison of Stochastic Mortality Models Using Data From England and Wales and the United States , 2009 .

[21]  Glen Barnett,et al.  2000 BEST ESTIMATES FOR RESERVES , 1998 .

[22]  R. Picken.  The age selection of mortality from tuberculosis in successive decades , 1939 .

[23]  N. Ryder.  The cohort as a concept in the study of social change. , 1965, American sociological review.

[24]  M. C. Sheps,et al.  A Technique for Analyzing Some Factors Affecting the Incidence of Syphilis , 1950 .

[25]  M. Sherris,et al.  The Application of Affine Processes in Multi-Cohort Mortality Model , 2015 .

[26]  Stuart A. Klugman,et al.  Loss Models: From Data to Decisions , 1998 .

[27]  Hong Chang,et al.  Model Determination Using Predictive Distributions with Implementation via Sampling-Based Methods , 1992 .

[28]  An approach to the analysis of claims experience in motor liability excess of loss reinsurance , 1972, ASTIN Bulletin.

[29]  S. Fienberg,et al.  Identification and estimation of age-period-cohort models in the analysis of discrete archival data , 1979 .

[30]  G. Casella,et al.  Explaining the Gibbs Sampler , 1992 .

[31]  James B. McDonald,et al.  Some Generalized Functions for the Size Distribution of Income , 1984 .