Forecast studies for financial markets using technical analysis

Abstract

The forecasting of the financial markets is an area of active research. Several methods have been studied with the aim of predicting future prices of financial instruments. In this paper, the use of technical analysis in forecasting and trading in the foreign exchange market is examined. Specifically, the application of neural networks in predicting future values of the US Dollar Vs British Pound is discussed. These predictions have been utilized in trading strategies and the returns obtained compared with those obtained using traditional technical indicators. The results show that neural networks are capable of predicting exchange rates with low values of mean square error. However, due to the high degree of complexity of the time series, it was not possible to obtain accurate enough predictions so as to generate consistent profits while trading. © 2005 IEEE.