Somasse, Gbetonmasse Blaise
In financial trading, emotion can often obstruct clear decision making. The goal of this project is to build a system which can overcome this by trading foreign currencies autonomously. Three systems were created: two relying on neural networks, and a third on pattern recognition of candlestick charts. A fourth system has been designed to allocate funds to the others using utility theory. Though the algorithms were not profitable, a powerful interface was built, connecting Python scripts to MetaTrader 4 for trading.
Worcester Polytechnic Institute
Interactive Qualifying Project
All authors have granted to WPI a nonexclusive royalty-free license to distribute copies of the work, subject to other agreements. Copyright is held by the author or authors, with all rights reserved, unless otherwise noted.