FX technical trading rules can be profitable sometimes!

详细信息    查看全文

• 作者：Nima Zarrabi^a ; ^{nzarra@essex.ac.uk} ; Stuart Snaith^b ; ^{ssnaith@uvic.ca} ; Jerry Coakley^a ; ^c ; ^{jcoakley@essex.ac.uk}
• 关键词：C12 ; C15 ; F31 ; G11 ; G14
• 刊名：International Review of Financial Analysis
• 出版年：2017
• 出版时间：January 2017
• 年：2017
• 卷：49
• 期：Complete
• 页码：113-127
• 全文大小：1181 K
• 卷排序：49

文摘

This paper investigates the profitability of technical trading rules in the foreign exchange market taking into account data snooping bias and transaction costs. A universe of 7650 trading rules is applied to six currencies quoted in U.S. dollars over the 1994:3–2014:12 period. The Barras, Scaillet, and Wermers (2010) false discovery rate method is employed to deal with data snooping and it detects almost all outperforming trading rules while keeping the proportion of false discoveries to a pre-specified level. The out-of-sample results reveal a large number of outperforming rules that are profitable over short periods based on the Sharpe ratio. However, they are not consistently profitable and so the overall results are more consistent with the adaptive markets hypothesis.