Description: 2011 SOA Enterprise Risk Management Symposium, Chicago. This paper suggests the use of a policy event-based loss estimation method using expert input and simulation for forecasting long duration expenditures: that is to say, there is an efficient time, which is random and observable to replace a physical asset.

Description: In this paper, we derive three important longevity Greeks on the basis of an extended version of the Lee-Carter model that incorporates stochastic volatility.

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.

Description: This paper presents a new methodology for obtaining fast algorithms and closed form solutions for pricing callable bonds in continuous time models. Results are directly applicable to regulatory scenario testing, immunization and market value accounting.

Description: In the copula literature there are many bivariate distribution families but very few higher dimensional ones. Moreover, most of these are difficult to work with. Some of the bivariate families can be extended to more dimensions but in general the construction of distribution functions with more than two variables is a difficult problem. This paper introduces a construction method that is straightforward to implement and can produce multivariate distribution functions of any dimension.

Description: Quantitative operational risk assessment is essentially based on stochastic scenario modeling of operational loss sequences. Given the lack of reliable historical data in most cases, mathematical methods should meet even stronger requirements in terms of results received and include specific features, namely: rare event analysis, uncertainty analysis and human factor analysis, etc. In this paper, the analytical-statistical simulation approach ASSA is considered as the most flexible approach to stochastic scenario loss sequence modeling and compared with Markov chain, fault/event tree and Monte Carlo, in terms of scenario loss sequence model adequacy and calculation cost. Probabilistic risk analysis software PRAISE and some applications are given. Presented at 2010 Enterprise Risk Management Symposium, Society of Actuaries, April 12-15, 2010.

Description: Given the lower incidence rate, use of decision tree techniques like Classification and Regression Tree CART in understanding credit or operational risk becomes quite challenging. A commonly adopted solution is biased sampling approach, where more weights are attached to bad customers to artificially hike the incidence or bad rate. While adopting this type of biased sampling approach, question of the best weight arises. This paper adopts an iterative approach in identifying the best weight. The best weight for bankruptcy BKO profiling problem in hand occurred when the incidence rate was around 50 percent where entropy reaches its maximum.

Description: This paper considers the Bayesian analysis of the generalized Poisson distribution GPD and the generalized Poisson regression GPR model. Markov chain Monte Carlo MCMC methods and random variety generation strategies such as adaptive rejection sampling ARS for log-concave densities and adaptive rejection Metropolis sampling ARMS are discussed and then utilized in order to advance the Bayesian analysis of these generalized Poisson models.