Description: This paper introduces actuaries to the concept of 'copulas,' a tool for understanding relationships among multivariate outcomes. A copula is a function that links univariate marginals to their full multivariate distribution. Recently, there has been a rapidly developing literature on the statistical properties and applications of copulas. This paper explores some of these practical applications, including estimation of joint life mortality and multidecrement models. In addition, the paper describes basic properties of copulas, their relationships to measures of dependence and several families of copulas that have appeared in the literature. This paper will also be useful to those who wish to use copulas for statistical inference. Statistical inference procedures are illustrated using insurance company data on losses and expenses. For this data, the paper shows how to fit copulas and describes their usefulness by pricing a reinsurance contract and estimating expenses for pre-specified losses.

Description: Shiu derives two formulas for calculating the probability of eventual ruin in a collective risk model. This paper implements one of the formulas by incorporating an algorithm to reduce round-off error due to convolution. It gives the results of the programs for some claim distributions.

Description: The dependent relationship between two lives is readily analyzed within the context of probability theory. This paper presents useful relationships between certain actuarial and probability functions and then explores measures of dependence. From Transactions of Society of Actuaries 1986, Vol. 38.

Description: Using data from a large insurance company, we construct a parametric model for the joint survival function based on a Hougaard copula function.

Description: This is a final report for a research project to create reference material that addresses the methods and techniques for quantifying margins for uncertainty in actuarial assumptions under a Principle-based statutory accounting framework.

Description: This paper studies the joint distribution of the time of ruin, the surplus immediately before ruin, and the deficit at ruin. The classical model is generalized by discounting with respect to the time of ruin. The authors show how to calculate an expected discounted penalty, which is due at ruin, and may depend on the deficit at ruin and the surplus immediately before ruin.

Description: This paper explores the use of linear programming as a tool to guide policyholders in getting the most value out of their combined insurance and investment programs. Concentrating on flexible premium universal life within the tax environment of the United States, several linear programming models are developed that can be used at the point of sale, to select the most cost effective policy from those available in the marketplace, and after issue to maximize the present value of future cash flows on the policy of an insured who is in ill-health. From Actuarial Research Clearing House 1991, VOL. 3.

Description: Reports generally display tables of data broken down by many variables. This paper presents a variation of the Whittaker-Henderson graduation method that may be used to graduate such multidimensional data. Followed by several discussion papers submitted

Description: This paper studies a smoothing spline Poisson regression model for the analysis of mortality data. From Actuarial Research Clearing House 2006, Vol. 1.

Description: In this paper, Bivariate Schuette graduation is applied to mortality rates among black male retired workers [Social Security data] at ages 70 to 83. Graduated values for various choices of the parameters are compared. The Schuette method is also adapted to calculate regression quantiles, providing upper and lower envelopes for the graduated values.