Description: This article pertains to the optimal asset allocation of surplus from an insurance company model. The insurance company is represented by a compound Poisson risk process which is perturbed by diffusion and has investments. The investments are in both risky and risk-free types of assets similar to stocks/real estate and bonds. The insurance company can borrow at a constant interest rate in the event of a negative surplus. Numerical analysis appears to show that an optimal asset allocation range can be estimated for certain parameters and compared with insurance data. Using a conservative method to minimize the probability of ruin, a reasonable optimal asset allocation range for a typical insurance company is about 4.5 to 8.2 percent invested in risky stock/real estate assets. An inequality and the exact solution are obtained for the pure diffusion equation. In addition, the asymptotic form of the ruin probability is shown to be a power function.

Description: Presentation at the 41st Actuarial Research Conference held on August 10-12, 2006 in Montreal, QC. This paper considers the problem of extending regime-switching models for asset returns to the situation with multiple asset markets. The conditional tail expectation CTE is highly sensitive to the assumed dependence structure between the regime models for the different markets. A Monte Carlo simulation is used to illustrate the cost of misspecification. The authors find that the model most commonly used by practitioners is worse than all of the alternatives that they investigate.

Description: This paper describes the use of the Moses software package to model variable annuities with multiple guarantees. Pricing and risk management are discussed.

Description: The behavior of extreme paths in the usual stochastic interest rate models is not nearly so plausible as the behavior of their expected values and variances. This paper proposes a new class of models that deliver more plausible extreme paths while preserving the usual expected value and variance behavior. Along the way we derive the closed form solution for the traditional mean-reverting lognormal process, including the drift compensation.