Description: Poisson regression models are commonly used in actuarial science to model count data e.g., the number of claims and covariates. The standard parametric formulation assumes that the response to covariates is constant across the population under study. Additionally, the shape of the posterior distribution is rather inflexible. In practice, however, the response to covariates may depend on a latent class and the posterior distribution may be highly nonparametric. This paper proposes a Bayesian nonparametric model in which a Dirichlet process prior on the regression coefficients leads to a clustering of observations into groups having the same response to covariates. This paper illustrates the utility of this approach with an example dataset of diabetes deaths.

Description: This abstract describes a paper that assesses the optimal manner to collapse a factor with many levels into one with a smaller number of levels.

Description: Feature article discussing the use of Bayesian Networks as predictive modeling tools.

Description: This presentation considers the impact of sampling variation on the calibration of stochastic mortality models.

Description: This paper attempts a model for a mix of DB and DC retirement plans that meets the dual objectives of predictability of employer cash contributions, and sufficient employee retirement benefit. From Current Pension Actuarial Practice in Light of Financial Economics Symposium, M-RS04-1.

Description: Exam study notes on credibility theory.

Description: This paper examines some new Bayesian models for loss reserving inspired by a consideration of some of the methods and techniques appearing in the traditional chain ladder literature. This includes a possibly order restricted hierarchical Bayesian model for the year over year development factors. The issues of reserve variability and ranges of reasonable reserve estimates are considered. A new Bornhuetter-Ferguson styled Bayesian method of estimate revision is also discussed. These Bayesian models can be implemented using Markov chain Monte Carlo MCMC methods in a variety of programming languages, and can serve as starting points for more advanced models. This paper describes their implementation in WinBUGS a specialized software program for MCMC simulations. Illustrative WinBUGS code is included.

Description: In this paper we propose a model that can be used to analyze correlated claims data. In the process we introduce a class of multivariate generalized Poisson distributions. Then we present posterior distributions of its parameters. These distributions are difficult to manipulate, so we employ Markov chain Monte Carlo methods to draw random numbers from posteriors and predictive distributions. We present a collection of EXCEL VBA functions and subroutines to perform the simulations.

Description: The purpose of this paper is to define a predictive model, based on known system attributes, for assessing risk associated with information systems.

Description: Book review of The Theory That Would Not Die: How Bayes’ Rule Cracked the Enigma Code, Hunted Down Russian Submarines and Emerged Triumphant from Two Centuries of Controversy by Sharon Bertsch McGrayne.