Optimal Dynamic Portfolio Risk Management

Abstract

Numerous econometric studies report that financial asset volatilities and correlations are time varying and predictable. Over the past decade, this knowledge has stimulated increasing interest in various dynamic portfolio risk control techniques. At present, the two basic types of risk control techniques—risk control across assets and risk control over time—are not implemented simultaneously, and there has been surprisingly little theoretical study of optimal dynamic portfolio risk management. This article fills this gap in the literature by formulating and solving the multiperiod portfolio choice problem. Using several datasets and performing out-of-sample simulations, the author demonstrates that it is optimal to simultaneously control portfolio risk across assets and over time. Specifically, the author shows that portfolios with risk control only across assets outperform equally weighted portfolios and that portfolios with risk control both across assets and over time outperform portfolios with risk control across assets only.