Description: This session of the SOA 2003 Washington, DC Spring Meeting covers the basics of stochastic modeling with focus on random number and scenario generation, steps in creating models, and verification of models and assumptions.

Description: From a session at the Spring meeting of the Society of Actuaries held in San Antonio, Texas, June 14-15, 2004 The panelists discuss the benefits of hedging variable annuity guarantees, the challenges they face and how they keep score.

Description: A contingency table is a set of counts or frequencies obtained by classifying observations in two or more different ways. Log-linear analysis is applied to disability terminations, linking them to variables such as sex, age, region and maximum benefit period. Application to risk classification and underwriting are outlines.

Description: Stochastic approximation is a powerful tool for sequential estimation of zero points of a function. In this paper, this methodology is defined and is shown to be related to a broad class of credibility formulae derived for the Exponential Dispersion Family. The authors further consider a Location Dispersion Family which is rich enough and for which no simple credibility formula exists. For this case, a Generalized Sequential Credibility Formula is suggested and an optimal stepwise gain sequence is derived.

Description: Design and future maintenance of an asset portfolio backing a new line of business is critical for proper asset and liability management for that business. Most portfolio optimization methods utilize linear or quadratic programming and require the user to specify the asset and liability attributes and cash flows into the program. The programmer must also supply an objective function to allow the program to find the optimal asset mix for the associated liabilities. Two problems with this approach are that the liability cash flows are fairly static and that one must frequently rebalance the portfolio. An alternative approach would be to develop a corporate model of both the assets and liabilities and incorporate various economic scenarios as input into the model. Asset strategies would be measured against a specific objective function that is calculated by the corporate model. Unfortunately, the majority of maximization algorithms available are very time consuming, and obtaining a reasonable portfolio mix becomes impractical. This paper overcomes this difficulty by using a very rapid optimization method from the chemical engineering profession called the Floppy Triangle. The paper includes an example of the use of this process to determine an optimal portfolio. It also discusses modifications of the algorithm that is required when the shorting of assets is not permitted.

Description: The purpose of this paper is to provide a method for calculating special contingency reserves for investment losses. The method is derived by first building a stochastic investment model and then utilizing its probabilistic structure. From Transactions of Society of Actuaries 1980, Vol. 32.

Description: We consider Libor market model and calibration process. We estimate some interest rate derivatives such as Caps and Swaptions. We use Monte Carlo methods with appropriate dynamics under spot measure and forward measure. Imposing martingale discretization property, the same quantities are estimated again to see a contrast or discrepancy in light of confidence interval. Based on algorithms discussed in the paper, “Arbitrage-free discretizations of lognormal forward Libor and swap rate models” by Glasserman, P and Zhao, X 2000, the algorithms are implemented in excel VBA. Starting with current market data such as FRAs and Bond Prices, we describe steps involved to estimate the price of interest rate derivatives such as Caps and Swaptions, which would be beneficial to the practitioners in the industry.

Description: This is a presentation from 43rd Actuarial Research Conference ARC, Regina, August 14–16, 2008. This talk will present the moment generating function of the discounted compound Poisson aggregate sums when the deflated claims are PH distributed and moment generating function of the discounted compound renewal sums.

Description: This research was funded by the Natural Sciences and Engineering Research Council of Canada [NSERC] Discovery Grant 36860–06. Loss reserving is one of the most challenging tasks facing actuaries. Numerous approaches have been developed to give reasonable estimates. Generalized linear models [GLMs] are becoming quickly popular statistical analysis methods to estimate loss reserves. However, most of these models are aggregate reserving methods based on loss development triangles, without using information with regard to the actual claims processes. This paper establishes a more sophisticated structural reserving method incorporating more detailed information, such as the premium exposure emergence pattern, the loss emergence pattern and the loss development pattern, within the framework of GLMs.

Description: The Risk Management Task Force polled the Risks & Rewards readers and asked the following questions related to Monte Carlo models: 1 How many scenarios do you run? 2 How did you determine the number? and 3 What confidence interval does your result have? The results are summarized in this article.