Interest Rate Arbitrage in Currency Baskets : Forecasting Weights and Measuring Risk

When constructing hedged interest rate arbitrage portfolios for basket currencies, two issues arise: first, how are the unknown future basket weights optimally forecasted from past exchange rate data? And, second, how is risk - in terms of the conditional variance of expected profits from the interest rate arbitrage portfolio - appropriately measured when the basket weights are time-varying? Answers to these questions are provided within a time-varying parameter modeling framework estimated through the Kalman filter. An empirical application is devoted to the experience of the Thai baht currency basket (January 1992-February 1997)