Description: Our previous article, “Introduction of Cashflow Matching Strategic Asset Allocation (SAA) Framework,” introduced the concept and included a real-world case study that developed a liability-driven investment strategy for a single line of business (LOB) for a life insurer. This article expands the framework and the case study from a single LOB to simultaneously optimize asset allocation across multiple LOBs. It discusses how this multi-line SAA framework can enable an insurance company to efficiently tag its asset holdings to each LOB, identify opportunities in the public and private asset markets to enhance risk-adjusted returns, and develop a deeper understanding of the trade-off between the degree of cashflow matching and risk-adjusted returns across LOBs.

Description: Risk management techniques do not include system science. System science requires that the uncertainty of the world is incorporated into the results. The articles explains the difference between risk and uncertainty. Learn how risk practices create an overoptimistic view of future outcomes.

Description: To address non-parallel movement in the interest rates the concept of key-rate duration was introduced in mid-'80s by Tom Ho. I will define certain key rate metrics, their variation used in the industry, how to calculate them, and an existing industry issue with how they are used. Additionally, I propose a novel technique to calculate key rates to mitigate the problem faced in the industry

Description: To address non-parallel movement in the interest rates the concept of key-rate duration was introduced in mid-'80s by Tom Ho. I will define certain key rate metrics, their variation used in the industry, how to calculate them, and an existing industry issue with how they are used. Additionally, I propose a novel technique to calculate key rates to mitigate the problem faced in the industry

Description: This presentation from the 2011 46th Actuarial Research Conference is about the Haezendonck-Goovaerts risk measure for extreme risks.

Description: This paper establishes both first-order and second-order asymptotics for the Frechet, Weibull and Gumbel cases.

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: There have been continual advances in the modeling of financial series but most are aimed at the pricing of derivatives. Different criteria are needed for development of scenarios for risk management. Some recent methods will be reviewed with an eye on risk-management applications, including using the simulated method of moments to parameterize multifactor models, fractional differencing and other methods to model series with persistent autocorrelation, and models to flatten out the volatility smile, such as jump-diffusion models. These methods will be illustrated with applications to inflation, interest rates, equity prices and exchange rates.

Description: The paper o ers a simple framework for ranking the common reinsurance struc- tures in practice with the theory of stochastic orders. The basic idea is to slice the space of reinsurance structures into groups by expected loss cost to facilitate the comparisons within the group and between groups. Given the standard risk aversion assumption in economics, a spectrum of reinsurance structures with the same expected loss cost can be compared analytically with one another and sequenced based on their risk coverages under the convex order. The paper then expands the dimension of the comparison to groups of reinsurance structures with di erent expected loss costs, which can be ranked under the increasing convex order and the usual stochastic order. As such, the paper maps out the ordering for the entire space of reinsurance structures and presents it in a ma- trix format for quick reference. The implication

Description: In a very general setting this paper describes a straightforward, non-parametric approach to aggregate "stand-alone" or marginal distributions with desired correlations without imposing additional assumptions on those marginals. We then develop two optimizations for the enterprise, one based on maximizing return on economic capital and the other based on a mean -semivariance efficient frontier from the investor point of view.