Metaheuristics application on a financial forecasting problem

Abstract

EDDIE is a Genetic Programming (GP) tool, which is used to tackle problems in the field of financial forecasting. The novelty of EDDIE is in its grammar, which allows the GP to look in the space of technical analysis indicators, instead of using prespecified ones, as it normally happens in the literature. The advantage of this is that EDDIE is not constrained to use prespecified indicators; instead, thanks to its grammar, it can choose any indicators within a pre-defined range, leading to new solutions that might have never been discovered before. However, a disadvantage of the above approach is that the algorithm's search space is dramatically larger, and as a result good solutions can sometimes be missed due to ineffective search. This paper presents an attempt to deal with this issue by applying to the GP three different meta-heuristics, namely Simulated Annealing, Tabu Search, and Guided Local Search. Results show that the algorithm's performance significantly improves, thus making the combination of Genetic Programming and meta-heuristics an effective financial forecasting approach.

Guided Fast Local Search for speeding up a financial forecasting algorithm

“…Results showed that all of these meta-heuristics provided significant improvements to the predictive performance of EDDIE-8. Smonou et al [2] also state that the GLS algorithm returned the best performance out of the three, significantly outperforming its opponents. However, a new issue arose, which was the algorithm's computational cost, as the GLS introduced a high number of additional fitness evaluations.…”

“…Due to its importance, an increasing number of investors and researchers have been attracted to investigate in the field [2]- [5]. In recent years, with the latest technological advancements, many financial forecasting tools, which take advantage of computational intelligence, have been released [6].…”

“…In recent years, with the latest technological advancements, many financial forecasting tools, which take advantage of computational intelligence, have been released [6]. A new version of a Genetic Programming (GP) financial forecasting tool called EDDIE-8 has been released [3], which is the extended version of EDDIE (Evolutionary Dynamic Data Investment Evaluator) [2], [3], [7], [8]. The new feature of EDDIE-8, as compared to its predecessor EDDIE-7, is the enriched and extended grammar of Genetic Decision Trees (GDTs) [3], [5], [7].…”

“…To be more specific, due to the significantly larger search space, it is easier for good solutions to be missed. One of the works that attempted tackling this issue was Smonou et al [2]. This work implemented three Meta-heuristics (Simulated Annealing (SA), Tabu Search (TS), and Guided Local Search (GLS)), which were applied to the period nodes of the GP trees.…”

“…Fast Local Search (FLS) [2], [3] is a heuristic, which has been proven helpful in improving the efficiency of GLS [10], as it has been found that it can significantly speed up the algorithm. Our goal thus in this paper is to implement FLS and combine it with GLS, as an attempt to reduce the computational time of GLS, while at the same time maintaining the algorithm's predictive performance at the same levels.…”