A fuzzy logic model for forecasting exchange rates

• First attempt at implementation of fuzzy logic in fundamental analysis.
• Comparison of the effectiveness of statistical and artificial intelligence models.
• High effectiveness of the forecast generated by the created model.
• Improved forecasting effectiveness of three currency pairs.

This article is devoted to the issue of forecasting exchange rates. The objective of the conducted research is to develop a predictive model with the use of an innovative methodology – fuzzy logic theory – and to evaluate its effectiveness in times of prosperity (years 2005–2007) and during the financial crisis (years 2009–2011).The model is based on sets of rules written by the author in the form of IF-THEN, where expert knowledge is stored. This model is the result of ten years of the author’s research on this issue.Empirically, this paper employs three currency pairs as experimental datasets: JPY/USD, GBP/USD and CHF/USD. From the model verification, it is demonstrated that refined processes are effective in improving the forecasting of exchange rate movements. The author’s created model is characterised by high efficiency. These studies are among the world’s first attempts to combine fundamental analysis with fuzzy logic to predict exchange rates.