Description: This paper develops a new technique, which allows the analyst to maximally use all thte historical interest rate information available in forecasting interest rates. Interest rates' daily change do not follow a stationary random process over long time periods. Instead, the process is comprised of stationary periods, each a few years long, with relatively short transitional periods between different stationary periods.

Description: A presentation at the Actuarial Research Conference in August 2006 in Montreal. This paper discusses the Basel VaR Value at Risk test of a bank's VaR model. It proposes an alternative, Quality Control of Risk Measures QCRM Hypothesis test. The alternative is more likely to reject an incorrect VaR model.

Description: From a session at a meeting of the Society of Actuaries held in San Diego, CA, June 22-23, 2000 Company representatives and state insurance department representatives addressed issues faced by companies in complying with the form-filing process. In addition, they offered suggestions and guidance in expediting policy form approvals.

Description: Chief actuary of the California Insurance Department provides excerpts from a paper he presented his personal views to the Actuarial Research Conference in October 1986. He states that the regulators need to be aware of a deteriorating financial condition of an insurer in time to protect its policyholders from the consequences of insolvency.

Description: Risk management theory prescribes a few legitimate reasons for management to hedge corporate revenues and reduce income volatility. In this paper we will use the oil tanker industry to illustrate some of these strategic considerations and how management can account for them in their decisions regarding whether and how much to hedge. Further work is needed to incorporate these considerations into a broader theoretical framework. Presented at 2010 Enterprise Risk Management Symposium, Society of Actuaries, April 12-15, 2010.

Description: Regularization, by means of trading restrictions, is an effective way to obtain sparse replicating portfolios. By including only the most relevant replicating instruments, the resulting portfolio is more efficient computationally, easier to interpret, and better able to approximate the liability on an out-of-sample basis. The instruments selected depend largely on their respective trading costs. This paper evaluates a number of alternative methods for specifying trading costs. We find that trading costs based on simple statistics of the instrument and liability cash flows are an effective choice in practice. Presented at 2010 Enterprise Risk Management Symposium, Society of Actuaries, April 12-15, 2010.