Many models have been built to predict the exchange rate.
Refenes [1] developed a constructive learning algorithm to
predict the exchange rate between U.S. dollar and the Deutsche
mark. Kuan and Liu [2] examined performance of feed-forward
Manuscript received March 6, 2007.
Hongxing LI was with Department of Statistics & Finance, University of
Science & Technology of China. Hefei, Anhui 230026 CHINA. He is now a
research associate at Electrical and Computer Engineering, University of
Michigan-Dearborn, MI 48128 USA (e-mail:hongxing@umd.umich.edu).
Zhaoben Fang is with the Department of Statistics & Finance, University of
Science & Technology of China, Hefei, Anhui 230026 CHINA (e-mail:
zbfang@ustc.edu.cn).
Dongming Zhao is with the Department of Electrical & Computer
Engineering, The University of Michigan-Dearborn, MI 48128 USA (e-mail:
dmzhao@umich.edu).
and recurrent neural. Diebold and Nason [3] investigated ten
weekly spot rates and did not find any significant difference in
both in-sample fit and out-of-sample forecasting across these
exchange rate series