Research on Automated Forex Trading System Based on BP Neural Network

Abstract:

Article Preview

This paper reports on an expert advisor for forex trading based on Back Propagation Neural Network (BPNN) in MetaTrader4 platform. A single hidden layer feedforward network was established for foreign exchange rate prediction. Trading rules based on the prediction results was designed and realized. Finally, we optimized the parameters according to the profitability performed on EUR/USD, GBP/USD currency pairs separately. The optimized results are able to achieve good results in the training series. In the test series, the strategies are consistently profitable for at least the first twenty days. It is concluded that the BPNN based model do have the ability to make profits from the experimental currency pairs for the period investigated.

Info:

Periodical:

Advanced Materials Research (Volumes 753-755)

Edited by:

Xiaoming Sang and Yun-Hae Kim

Pages:

3080-3083

DOI:

10.4028/www.scientific.net/AMR.753-755.3080

Citation:

L. Meng and Y. Sun, "Research on Automated Forex Trading System Based on BP Neural Network", Advanced Materials Research, Vols. 753-755, pp. 3080-3083, 2013

Online since:

August 2013

Authors:

Export:

Price:

$35.00

In order to see related information, you need to Login.

In order to see related information, you need to Login.