Extreme learning machine based financial prediction

• Main Author: Huang, Fei.
• Other Authors: Huang Guangbin
• Format: Final Year Project
• Language: English
• Published: 2011
• Subjects: DRNTU::Engineering::Electrical and electronic engineering::Computer hardware, software and systems
• Online Access: http://hdl.handle.net/10356/45819
• Institution: Nanyang Technological University

The stock market prediction is one the hottest topics since it could yield significant profits by successful prediction of a stock’s trading signals or stock’s future trends. The recent researches pay more attention to stock tendency prediction, which involved many technological methods. Various machine learning approaches have been proposed for stock prediction. However, due to the complexity and randomness of stock market, stock trend prediction issues remain unsolved now. In this report, a new learning algorithm based financial prediction mechanism called Extreme Learning Machine (ELM) based Financial Prediction System is presented. This system integrates the ELM stock trend prediction with technical analysis to generate trading signal and calculate the profit and loss. In our proposed system, we firstly search for the trading signals using technical indicators, and then apply ELM trend prediction results to filter the trading signals. Experimental results demonstrate the positive contribution of ELM (the acceptable prediction accuracy) and the ELM based financial prediction system (profit generation).