Enhancing technical analysis in the FOREX market using neural networks

Chun Chung Chan, Kean Teong Foo

Research output: Chapter in book / Conference proceedingConference article published in proceeding or bookAcademic researchpeer-review

16 Citations (Scopus)

Abstract

Copious test by countless professionals have proven Technical Analysis to be, at best, break-even tools, even with the finest money management techniques. Those who use only Technical Analysis in actual trades find out very painfully what whipsaws, and false breakouts are. The simple mathematical explanation for this is that Technical Analysis, as introduced by Wilder, Lane, etc., are linear, monovariate computation routines. This means that Technical Analysis is not designed to deal with non-uniform periodic, and discontinuous functions. To manage these inadequacies, one employs a neural network. The simple network described in this paper predicts technical indicators, and generates trading signals before regular technical indicators do. This gives one the opportunity to enter, and exit trades before the crowd. Tests, and actual trades, have shown that most of the time, one or two days, is all the advantage one needs.
Original languageEnglish
Title of host publicationIEEE International Conference on Neural Networks - Conference Proceedings
PublisherIEEE
Pages1023-1027
Number of pages5
Publication statusPublished - 1 Dec 1995
EventProceedings of the 1995 IEEE International Conference on Neural Networks. Part 1 (of 6) - Perth, Australia
Duration: 27 Nov 19951 Dec 1995

Conference

ConferenceProceedings of the 1995 IEEE International Conference on Neural Networks. Part 1 (of 6)
CountryAustralia
CityPerth
Period27/11/951/12/95

ASJC Scopus subject areas

  • Software

Cite this