EDDIE beats the bookies

Abstract

SUMMARYInvestment involves the maximisation of return on one's investment whilst minimising risk. Good forecasting, which often requires expert knowledge, can help to reduce risk. In this paper, we propose a genetic programming-based system, EDDIE (Evolutionary Dynamic Data Investment Evaluator), as a forecasting tool. Genetic programming is inspired by evolution theory, and has been demonstrated to be successful in other areas. EDDIE interacts with the users and generates decision trees, which can also be seen as rule sets. We argue that EDDIE is suitable for forecasting because apart from utilising the power of genetic programming to efficiently search the space of decision trees, it allows expert knowledge to be channelled into forecasting, and it generates rules which can easily be understood and verified. EDDIE has been applied to horse racing and achieved outstanding results. When experimented on 180 handicap races (real data) in the UK, it out-performed other common strategies used in horse race betting by great margins. The idea was then extended to financial forecasting. When tested on historical S&P-500 data EDDIE achieved a respectable annual rate of return over a three and a half year period. While luck may play a part in the success of EDDIE, our experimental results do indicate that EDDIE is a tool which deserves more research. © 1998 John Wiley & Sons, Ltd. key words: finance; forecasting; genetic programming; horse racing; investment INTRODUCTION Financial investment is concerned with risking money in order to gain more money. The objective is to minimise risk and maximise Return On Investment (ROI). Information can often help us to reduce risk. For example, to predict the future price of a share, useful information include; its past and present prices, price-earning ratios, price-to-book values, dividends, market indices, who said what in the public domain, etc. The aim of our research is to build tools to help investors to make best use of the information available to them. Such tools should improve the productivity of the users by allowing them to examine more sets of rules in less time. To some users, such tools could help them to do what they did not have the knowledge to do without such tools.Many factors could directly or indirectly affect the future price of an investment. Such factors are often inter-related, which adds to the difficulty of analysis. The combinatorial explosion problem prevents one from examining combinations of all the factors and their possible ways of interaction. We have developed EDDIE

GP algorithm versus hybrid and mixed neural networks

“…This has particularly been the case in the area of probabilistic models as GP has been incorporated with schema theories and Markov chain models. A variety of Genetic Programming applications is shown in the papers below: Wincler (2004), Wincler et al (2004a, b), Madar et al (2004Madar et al ( , 2005, Willis et al (1997), Tsang et al (1998), Fukunaga and Stechert (1998) and Werner and Fogarty (2001).…”

EDDIE-Automation, a decision support tool for financial forecasting

“…The first implementation of EDDIE, called EDDIE-1, was applied to horse racing [21]. It's performance was promising but limited, due to the lack of data in horse racing at the time.…”

Combining Ordinal Financial Predictions with Genetic Programming

“…FGP (Financial Genetic Programming) is a genetic programming implementation specialized for financial forecasting. It is built as a forecasting tool under the EDDIE project [16]. In this paper, we shall focus on its application in combining individual expert predictions in order to generate better predictions.…”