Genetic Algorithms and Neural Networks for Stock Trading Prediction and Technical Signal Optimization

Abstract

There has been a growing interest in using trading rules that are based on the predicted directions of stock price movement. However, transaction costs have been ignored by most studies. In reality, a profitable trade cannot be guaranteed if profits from trading are not large enough to compensate for the costs of performing the transaction. This paper introduces the use of genetic algorithms for finding alternative directions that maximize trading profitability. These directions are used along with selected technical indicator inputs to model two artificial neural networks. The experimentations are tested on three stocks across different market industries. The overall results indicate that profits from trading guided by the neural networks are higher than those guided by the benchmarks, namely the MACD indicator, ARIMA, random walk, and the buy-and-hold.

Department(s)

Engineering Management and Systems Engineering

Document Type

Article - Conference proceedings

Document Version

Citation

File Type

text

Language(s)

English

Rights

© 2024 Decision Sciences Institute, All rights reserved.

Publication Date

01 Dec 2002

This document is currently not available here.

Share

 
COinS