Currency risk management decision involves deciding on when, how much and what hedging instrument (i.e., currency futures or options) should be used to hedge its risk exposure with the base currency. Intuitively the accuracy in forecasting the direction and magnitude of future exchange rate movements is central to currency risk management decision-making process. This research investigates the predictive performance of a hybrid multivariate model, using multiple macroeconomic and microstructure of foreign exchange market variables. Conceptually, the proposed system combines and exploits the merit of adaptive learning artificial neural network (ANN) and intuitive reasoning (fuzzy-logic inference) tools. An ANN is employed to forecast a foreign exchange rate movement which is followed by the intuitive reasoning of multi-period foreign currency returns using multi-value fuzzy logic for foreign currency risk management decision-making. Empirical tests with statistical and machine learning criteria reveal plausible performance of its predictive capability. (c) 2006 Elsevier B.V. All rights reserved.
|Pages (from-to)||942 - 951|
|Number of pages||10|
|Publication status||Published - 2007|