Evolving Trading Signals at Foreign Exchange Market
Paper examines the merit of evolutionary algorithms to generate trading signals for trading decisions at financial markets. We focus on foreign-exchange market. It is among the largest financial markets. “Technical” traders base their decisions on a set of technical rules evolved from past market activity. We employ a genetic algorithm to learn a set of profitable trading rules considering transaction costs; each rule generates a ‘buy’, ‘hold’, or ‘sell’ signal using moving average technical rule. We empirically evaluate our approach using exchange rates of four major currency pairs over the period 2000 to 2015. Performance evaluation on out-of-sample data indicates that our approach is able to provide acceptably high returns on investment. Comparison with exhaustive search proves convincing performance of our approach.
KeywordsTrading rules Forex market Excess returns Evolutionary algorithms
- 5.Lissandring, M., Daly, D., Sornette, D.: Statistical testing of DeMark Technical Indicators on Commodity Futures (2016). Early review paperGoogle Scholar
- 7.Menkhoff, L., Schlumberger, M.: Persistent profitability of technical analysis on foreign exchange markets? PSL Q. Rev. 48(193), 189–215 (2013)Google Scholar
- 15.Gallo, C.: The Forex market in practice: a computing approach for automated trading strategies. Int. J. Econ. Manag. Sci. 3(169), 1–9 (2014)Google Scholar
- 18.Koza, J.R.: Introduction to genetic programming. Adv. Genet. Program. 1, 21–45 (1994)Google Scholar
- 20.Pelusi, D., Tivegna, M., Ippoliti, P.: Intelligent algorithms for trading the euro-dollar in the foreign exchange market. In: Corazza, M., Pizzi, C. (eds.) Mathematical and Statistical Methods for Actuarial Sciences and Finance, pp. 243–252. Springer, Cham (2014). doi: 10.1007/978-3-319-02499-8_22 Google Scholar