Title :
ATA: the Artificial Technical Analyst. Building intra-day market strategies
Author_Institution :
DIEM, Genoa Univ., Italy
Abstract :
A trading system based on artificial neural networks is presented: the Artificial Technical Analyst (ATA) is part of a project focusing on the capabilities of variants of self-organising maps (SOMs) as pattern recognition tools, as well as on their perspective use as forecasters. The basic idea is to take advantage of the analogies shared by SOMs and human technical traders, i.e. their common search for known patterns through which they extrapolate useful knowledge to forecast future price movements. The project is depicted in terms of its guidelines, discussing experimental results over S&P500 intra-day futures market contract prices
Keywords :
commodity trading; contracts; data mining; electronic trading; extrapolation; forecasting theory; pattern recognition; self-organising feature maps; ATA; Artificial Technical Analyst; S&P500 intra-day futures market contract prices; artificial neural networks; data mining; future price movement forecasting; high-frequency data; intra-day market strategies; known pattern searching; pattern recognition; project guidelines; self-organising maps; stock market; topology-representing networks; trading system; useful knowledge extrapolation; Chaos; Economic forecasting; Engines; Gaussian processes; Guidelines; Linearity; Network topology; Pattern analysis; Pattern recognition; Space technology;
Conference_Titel :
Knowledge-Based Intelligent Engineering Systems and Allied Technologies, 2000. Proceedings. Fourth International Conference on
Conference_Location :
Brighton
Print_ISBN :
0-7803-6400-7
DOI :
10.1109/KES.2000.884150
