Paper Details: Downloads: 599
Serial Number: P1121022077
Title: Stock Market Indices Prediction via Hybrid Sigmoid Diagonal Recurrent Neural Networks and Enhanced Particle Swarm Optimization
Authors: Mahmoud A. Fakhreldin and Tarek Aboueldahab
Abstract: Stock market analysis is one of the most important and hard problems in finance analysis field. Recently, the usage of intelligent systems for stock market prediction has been widely established. This paper investigates the development of novel reliable and efficient technique to model the behavior of stock markets indices. The Sigmoid Diagonal Recurrent Neural Network (SDRNN) based Enhanced Particle Swarm Optimization (EPSO) algorithm is proposed, and used for Nasdaq-100 index and S&P 500 stock index analysis where the parameters of SDRNN are optimized using EPSO. Experimental results show that our proposed technique SDRNN- EPSO could represent the stock indices behavior very accurately and can provide the required level of performance better than using Basic-PSO.
Keywords: Sigmoid Diagonal Recurrent Neural Networks, Enhanced Particle Swarm Optimization, Time Series Prediction, Stock Market
Journal/Conference: International Journal of Artificial Intelligence and Machine Learning
Volume: 10
Issue: 1
Submission Date: 5/24/2010 12:00:00 AM
Review Date: 7/12/2010 12:00:00 AM
Publishing Date: 9/15/2010 12:00:00 AM
Article Downloads: 599
Download:

Facebook