Description: In this Chapter we present a novel way for estimating aggregated EC figures based on Bayesian copula estimation. Contrary to the classic approach of using a single inter-risk- correlation matrix we derive a probability distribution of possible correlation matrices that enables us to tackle the important issue of parameter uncertainty. Presented at 2010 Enterprise Risk Management Symposium, Society of Actuaries, April 12-15, 2010.

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: 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: This paper develops credibility using a longitudinal data framework. In a longitudinal data framework, one might encounter data from a cross-section of risk classes towns with a history of insurance claims available for each risk class. For the marginal claims distributions, we use generalized linear models, an extension of linear regression that also encompasses Weibull and Gamma regressions. Copulas are used to model the dependencies over time specifically, this paper is the first to propose using a t-copula in the context of generalized linear models. The t-copula is the copula associated with the multivariate t-distribution like univariate t-distributions, it seems especially suitable for empirical work. Moreover, we show that the t-copula gives rise to easily computable predictive distributions that we use to generate credibility predictors. Like Bayesian methods, our copula credibility prediction methods allow us to provide an entire distribution of predicted claims, not just a point prediction.