Integrated dynamic models for hedging international portfolio

Abstract

We develop scenario-based stochastic programming models for hedging the risks of international portfolios using options. The models provide an increasing level of integration in managing market and foreign exchange (FX) risks. We start with a single-stage model with currency options for selective hedging of FX risks, while market risk is addressed only through diversi cation. Stock options are then added to hedge market risks. Finally, using an innovative method to price quantos on the scenario tree underpinning the stochastic program, we develop integrated models with stock options or quantos and currency options. We extend the models to multi-stage settings. Backtesting the models on market data over a 14-year period that encompasses the global financial crisis of 2008, we demonstrate their effectiveness as they take increasingly integrated views of risk management. Simultaneous hedging of market and FX risks using stock and currency options yields the best ex-post performance of international portfolios. The di erences are economically signi cant, and statistical significance is established through rigorous hypothesis testing. The models are particularly e ective during the crisis. Test results show that two-stage models outperform their single-stage counterparts, regardless of the hedging strategy.