Bridging the Gap between Strategic Allocation and Investment Risk

For many institutional investors, there is a potential inconsistency between models used for long term strategic asset allocation and investment risk management. Investment risk models, often calibrated with a shorter history spanning five to fifteen years, could provide misleading results when used for portfolio construction decisions, which usually take into account longer term asset characteristics spanning multiple business cycles. In this article, we propose a methodology inspired by change of measure and importance sampling to address this challenge. We show that it is possible to reflect long term asset characteristics in the simulated scenarios generated by a risk system, creating a better alignment between the risk and long term asset allocation models. Our methodology allows institutional investors to better utilize their existing simulation from risk models for portfolio allocation, sensitivity analysis, stress testing and other portfolio applications