Exploiting trends in the foreign exchange markets

Abstract

We offer further evidence on the relevance of technical trading in exchange-rate markets using daily data for 95 currencies against the US dollar. To that end, we investigate the profitability of a simple technical trading rule based on Taylor's (1980) price trend model, generating optimal one-step-ahead forecasts of returns using genetic algorithms. These trading rules, that bear similarity to the popular trading rules based on moving averages, overcome the buy-and-hold strategy in 25 of 39 cases where trends are detected, even in the presence of transaction costs.

Keywords:

• exchange rates
• price trend model
• genetic algorithms
• trading rules

JEL Classification:

• C53
• F31
• G14

Acknowledgements

The authors thank Ethan Ilzetzki for providing us with the updated database on exchange-rate arrangements. The authors gratefully acknowledge financial support from the Spanish Ministry of Science and Innovation, through the research projects ECO2008-05565 and ECO2010-21318. Adrián Fernández-Pérez also acknowledges the financial support from INNOVA CANARIAS 2020.

Notes

¹This period differs between series depending on data availability.