Genetic algorithms and Darwinian approaches in financial applications: A survey

doi:10.1016/j.eswa.2015.06.001

Expert Systems with Applications

Volume 42, Issue 21, 30 November 2015, Pages 7684-7697

https://doi.org/10.1016/j.eswa.2015.06.001 Get rights and content

Highlights

•
A survey of evolutionary computation applied to finance.
•
GAs, GP, LCS, MOEAS, EDAS and co-evolution approaches are covered.
•
This new version makes a revision of similar surveys to refine the scope.
•
An analysis of past and new references allowed determines changes of interest in the field.
•
The unexplored combinations of problems and solution methods are indicated and discussed.

Abstract

This article presents a review of the application of evolutionary computation methods to solving financial problems. Genetic algorithms, genetic programming, multi-objective evolutionary algorithms, learning classifier systems, co-evolutionary approaches, and estimation of distribution algorithms are the techniques considered. The novelty of our approach comes in three different manners: it covers time lapses not included in other review articles, it covers problems not considered by others, and the scope covered by past and new references is compared and analyzed. The results concluded the interest about methods and problems has changed through time. Although, genetic algorithms have remained the most popular approach in the literature. There are combinations of problems and solutions methods which are yet to be investigated.

Introduction

Financial systems have a remarkable impact on society. For example, indexes such as the DJI (Dow Jones) or the S&P500 are commonly used to measure economic buoyancy. Financial markets provide a comfortable way to generate profit from current wealth and protect investors from adverse effects like inflation. Some authors (Cerda, 2005) have predicted an impending crisis of social security systems, a scenario where personal investment and pension funds will be of crucial importance.

On the other hand, the idea to exploit financial markets to make profit has always been in the mind of investors. For example, Lo, Mamaysky, and Wang (2000) made a description of the technical analysis (TA) charting approach, which has been used by investors for decades for this end. TA holds that security price history summarizes all the information available about a particular asset. Therefore, it is possible to find patterns, and exploit them for profit.

There are three main reasons to use evolutionary computation approaches (such genetic algorithms) in financial applications:

•
The limited reasoning hypothesis (LRH) (Lakemeyer, 1994) is a realistic assumption to complete the efficient market hypothesis (EMH) (Finger & Wasserman, 2004). This means all investors will make the best decision possible, but this one will depend on their computation power and ability to process financial data.
•
There is a massive amount of financial data available like never before in history.
•
The computational power available is vast and increases continuously.

EMH is based on the assumption investors are rational, therefore all of them will make the optimal decision. In the definition presented by Malkiel and Fama (1970), there is no room for advantages because the information is available to all the investors at the same time. The limited reasoning hypothesis (LRH) does not contradict the rationality assumption, but realizes the investor might fail to foresight all the possible options available.

Section snippets

Genetic algorithms and Darwinian approaches

The number of approaches proposed for financial applications is vast. The scope of the survey must be delimited to find proper conclusions. Genetic algorithms (GAs) are included in the evolutionary computation (EC) field. Nevertheless, EC includes other methods inspired by nature, culture, and language. Memetic algorithms are an example of culture-inspired evolutionary algorithms. GAs are a simplified version of Darwinian evolution. The schemata theorem shows how the best solutions pass their

Refining the scope of this work

The existence of surveys with similar approaches to this one seemed plausible. Therefore, there is the possibility to cover references already cited in similar works. For this reason, the scope of our paper was delimited by the review of surveys with related subjects than this one. The novelty of our approach comes in three different manners: this work covers time lapses not included in other review articles, it covers problems not considered by others, and the scope covered by past references

Description of financial applications

This section makes a description of each application found in the literature. The review of applications is presented classified by problem.

Discussion and open problems

Results of the review are summarized in Table 3, and Fig. 1, Fig. 2. The surveys used to define the scope of this review were not considered. Table 3 presents the percentage of references which treated a particular problem and used a particular solution method. Open research areas can be identified when a white space is present. White spaces indicate the specific problem has not been solved with a specific solution method. Fig. 1 summarizes total percentage of references per solution method.

Conclusions

This work presented a review of evolutionary computation to financial applications. The scope was limited to Darwinian approaches. Darwinian approaches are population-based approaches where selection of the fittest is implemented to find solutions. Genetic algorithms, genetic programming, learning classifier systems, multi-objective evolutionary algorithms, co-evolutionary algorithms and evolutionary estimation of distribution algorithms were the solution methods considered in this review.

The

Acknowledgements

The first and second authors are with the Research Group with Strategic Focus on Intelligent Systems of the National School of Engineering and Sciences at the Tecnológico de Monterrey, and gratefully acknowledge its support.

The first author thanks the Consejo Nacional de Ciencia y Tecnología (CONACyT) for its financial support through the PNPC program.

References (126)

K. Andriosopoulos et al.
Performance replication of the spot energy index with optimal equity portfolio selection: Evidence from the UK, US and Brazilian markets
European Journal of Operational Research
(2014)
R.d.A. Araújo et al.
A morphological-rank-linear evolutionary method for stock market prediction
Information Sciences
(2013)
G.S. Atsalakis et al.
Surveying stock market forecasting techniques–part II: Soft computing methods
Expert Systems with Applications
(2009)
B. Back et al.
Neural networks and genetic algorithms for bankruptcy predictions
Expert Systems with Applications
(1996)
Y.-H. Chang
Adopting co-evolution and constraint-satisfaction concept on genetic algorithms to solve supply chain network design problems
Expert Systems with Applications
(2010)
J.-S. Chen et al.
Constructing investment strategy portfolios by combination genetic algorithms
Expert Systems with Applications
(2009)
A.H. Gandomi et al.
Multi-stage genetic programming: A new strategy to nonlinear system modeling
Information Sciences
(2011)
W.D. Hillis
Co-evolving parasites improve simulated evolution as an optimization procedure
Physica D: Nonlinear Phenomena
(1990)
J.H. Holmes et al.
Learning classifier systems: New models, successful applications
Information Processing Letters
(2002)
Y. Hu et al.
Stock trading rule discovery with an evolutionary trend following model
Expert Systems with Applications
(2015)

K.J. Kim et al.

Genetic algorithms approach to feature discretization in artificial neural networks for the prediction of stock price index

Expert Systems with Applications

(2000)

G. Lakemeyer

Limited reasoning in first-order knowledge bases

Artificial Intelligence

(1994)

F. Lin et al.

Novel feature selection methods to financial distress prediction

Expert Systems with Applications

(2014)

K. Lwin et al.

A learning-guided multi-objective evolutionary algorithm for constrained portfolio optimization

Applied Soft Computing

(2014)

E. Ngai et al.

The application of data mining techniques in financial fraud detection: A classification framework and an academic review of literature

Decision Support Systems

(2011)

V. Ravi et al.

Soft computing system for bank performance prediction

Applied Soft Computing

(2008)

A. Adebiyi et al.

Portfolio selection problem using generalized differential evolution 3

Applied Mathematical Sciences

(2015)

R. Almgren et al.

Optimal execution of portfolio transactions

Journal of Risk

(2001)

A. Bahrammirzaee

A comparative survey of artificial intelligence applications in finance: Artificial neural networks, expert system and hybrid intelligent systems

Neural Computing and Applications

(2010)

D. Bernardo et al.

A genetic type-2 fuzzy logic based system for financial applications modelling and prediction

Bolsa Mexicana de Valores (2015). Notas sobre...

R.A. Cerda

On social security financial crisis

Journal of Population Economics

(2005)

S.-H. Chen

Agent-based computational macro-economics: A survey

A.-P. Chen et al.

Using extended classifier system to forecast S&P futures based on contrary sentiment indicators

G. Chen et al.

A hybrid of adaptive genetic algorithm and pattern search for stock index optimized replicate

Y. Chen et al.

Generating trading rules on the stock markets with robust genetic network programming using variance of fitness values

S.-H. Chen et al.

Option pricing with genetic algorithms: A second report

M.B. da Costa Moraes et al.

Evolutionary models in cash management policies with multiple assets

Economic Modelling

(2014)

R. de Araujo et al.

An evolutionary morphological approach for financial time series forecasting

K. Deb et al.

A fast elitist non-dominated sorting genetic algorithm for multi-objective optimization: NSGA-II

Lecture Notes in Computer Science

(2000)

R.F. de Brito et al.

Comparative study of FOREX trading systems built with SVR+ GHSOM and genetic algorithms optimization of technical indicators

C.L. del Arco-Calderón et al.

Forecasting time series by means of evolutionary algorithms

J.P. Donate et al.

Evolutionary optimization of sparsely connected and time-lagged neural networks for time series forecasting

Applied Soft Computing

(2014)

C. Ferreira

Gene expression programming: A new adaptive algorithm for solving problems

Complex Systems

(2001)

M. Finger et al.

Approximate and limited reasoning: Semantics, proof theory, expressivity and control

Journal of Logic And Computation

(2004)

R. Franke

Coevolution and stable adjustments in the cobweb model

Journal of Evolutionary Economics

(1998)

Garcia-Almanza, A. L., & Tsang, E. P. (2006). Forecasting stock prices using genetic programming and chance discovery....

S. García et al.

Multiobjective algorithms with resampling for portfolio optimization

Computing and Informatics

(2014)

A. Gaspar-Cunha et al.

Self-adaptive MOEA feature selection for classification of bankruptcy prediction data

The Scientific World Journal

(2014)

P. Gerard et al.

YACS: A new learning classifier system using anticipation

Soft Computing

(2002)

P. Ghosh et al.

Financial time series forecasting using agent based models in equity and FX markets

G. Giulioni et al.

Building artificial economies: From aggregate data to experimental microstructure. A methodological survey

D. Goldberg

(1989)

S. Goonatilake et al.

Genetic-fuzzy systems for financial decision making

J. Grefenstette et al.

Genetic algorithms for the traveling salesman problem

P. Gupta et al.

Asset portfolio optimization using support vector machines and real-coded genetic algorithm

Journal of Global Optimization

(2012)

S.B. Hamida et al.

Applying dynamic training-subset selection methods using genetic programming for forecasting implied volatility

Computational Intelligence

(2014)

G.R. Harik et al.

The compact genetic algorithm

IEEE Transactions on Evolutionary Computation

(1999)

G.R. Harik et al.

Linkage learning via probabilistic modeling in the extended compact genetic algorithm (ECGA)

A. Hirabayashi et al.

Optimization of the trading rule in foreign exchange using genetic algorithm

Cited by (134)

Adaptive locally weighted support vector algorithm with asymmetrically parametric insensitive/margin model
2024, Knowledge-Based Systems
In many fields such as finance and electrical load, locally weighted support vector regression (LWSVR) method can achieve better prediction results by constructing different sub-models for each prediction target. However, it selected a fixed and single nearest neighbor parameter of $k_{global}$ for each query, and adopted the traditional $k$ -nearest neighbor (KNN) algorithm, which is suboptimal. The weights in LWSVR assigned to the neighboring sample points also affected the accuracy of the model predictions. In this paper, we propose an adaptive locally weighted support vector regression with asymmetrically parametric insensitive/margin model, called Asy- $p a r$ - $v$ -ALWSVR. Since the selection of similar sample for query points is significant. We propose a novel locally $q$ -neighbour selection algorithm that adaptively selects the most suitable number of neighbours for each query point based on Mahalanobis distance, rather than using a fixed $k_{global}$ . This significantly improves prediction accuracy. The weights of neighbours for each query point in Asy- $p a r$ - $v$ -ALWSVR is based on the distance within the kernel-induced feature space. This effectively discerns the importance of samples during the training process. In particular, Asy- $p a r$ - $v$ -ALWSVR can handle skewed noise with heteroscedastic structure in regression problems. Finally, we evaluate the proposed model on a synthetic sinc dataset and 14 real datasets. The results demonstrate that Asy- $p a r$ - $v$ -ALWSVR outperforms six state-of-the-art SVR-based models.
An evolutionary Chameleon Swarm Algorithm based network for 3D medical image segmentation[Formula presented]
2024, Expert Systems with Applications
Developing 3D Convolutional Neural Networks (CNNs) for medical image segmentation is challenging due to the limited number of available labeled medical images and computational resources. Effective design of a 3D CNN requires a better structure and appropriate hyperparameters. Manually designing a neural network and choosing appropriate hyperparameters requires extensive knowledge, a time-consuming and challenging process, resulting in complex and over-parameterized networks. Neuroevolution is an automatic and effective approach to finding an optimal network structure and hyperparameters. This paper proposes an evolutionary based method using Chameleon Swarm Algorithm (CSA) to develop 3D CNN networks for volume segmentation. We also design a search space to build optimal 3D block-based encoder–decoder structure networks for medical image segmentation. We evaluated the proposed method on two different datasets (CT Spleen and MRI Heart) and compared the results of the obtained model with state-of-the-art models. Furthermore, the comparative results show that our proposed method yields 3D CNN networks with superior performance and fewer parameters than other state-of-the-art models utilized in 3D medical image segmentation.
Chaos measure dynamics in a multifactor model for financial market predictions
2024, Communications in Nonlinear Science and Numerical Simulation
To answer the question if chaos changes over time, we apply rolling windows to wavelet-denoised logarithmic S&P500 returns (2000–2020) and calculate consecutive chaos measures (e.g., Hurst-, maximum Lyapunov exponent or sample entropy). We state time-variation of the chaos measure series, indicating chaos instability or inherent chaotic time variations of the underlying (hyper-)chaotic deterministic S&P500 return system. Moreover, we intend to deploy these chaos measure series as predictors for the denoted financial series. An optimised selection of these series is applied as input features for a dynamic factor model realised as deep learning multilayer perception neural network to predict the original S&P500 price and return series out-of-sample. The approach is validated by performance metrics (e.g., explained variance score) and the residuals are shown to be non-autocorrelated and ∼iid. Finally, we compare the results with selected base or benchmark models (e.g., autoregressive models). Thus, the approach provides a novel multifactor model for practical market price predictions from a dynamical (inherent) system-based view.
Artificial intelligence based prediction of optimum operating conditions of a plate and fin heat exchanger under uncertainty: A gray-box approach
2023, International Journal of Heat and Mass Transfer
This study is based on gray-box modeling for the prediction of optimum mass flow rates of inlet streams of a Plate and Fin Heat Exchanger under uncertainty. A first principle model of the Plate and Fin Heat Exchanger was developed in Aspen Exchanger Design and Rating. Genetic algorithm was integrated with the first principle model to achieve the highest possible exit temperature of the cold process stream under uncertainty. A dataset of uncertain process conditions and their corresponding optimum inlet flow rates derived through the first principle-genetic algorithm integration was used to develop an artificial neural networks model. The artificial neural networks model was then integrated with the first principle model by replacing the genetic algorithm to form a novel gray-box framework. The proposed gray box model, i.e., artificial neural networks and first principle integration, achieved a higher effectiveness and higher outlet temperature than those derived through the straight run first principle model. The performance of the gray box framework was also comparable to the first principle integrated genetic algorithm approach and significantly minimized the computation time needed for estimating the optimum process conditions. First principle integrated genetic algorithm approach enhanced the effectiveness of the straight run model by 3.05%. Performance of the proposed gray-box model was comparable to the integrated framework of the genetic algorithm with the first principle model but was significantly faster. The developed artificial neural networks model was employed as a surrogate in Sobol and FAST sensitivity analysis framework to identify the impact of input variables on output variables. The proposed gray box based method enhanced the capability of the plate and fin heat exchanger to recover energy from the process stream and its robustness to cope with uncertainty. The proposed approach is suitable for real-time application and would contribute to laying a foundation for petroleum refinery 4.0.
Financial applications of machine learning: A literature review
2023, Expert Systems with Applications
This systematic literature review analyses the recent advances of machine learning and deep learning in finance. The study considers six financial domains: stock markets, portfolio management, cryptocurrency, forex markets, financial crisis, bankruptcy and insolvency. We provide an overview of previously proposed techniques in these areas by examining 126 selected articles across 44 reputed journals. The main contributions of this review include an extensive examination of data characteristics and features used for model training, evaluation of validation approaches, and model performance addressing each financial problem. A systematic literature review methodology, PRISMA, is used to carry out this comprehensive review. The study also analyses bibliometric information to understand the current status of research focused on machine learning in finance. The study finally points out possible research directions which might lead to new inquiries in machine learning and finance.
Neural network systems with an integrated coefficient of variation-based feature selection for stock price and trend prediction
2023, Expert Systems with Applications
Stock market forecasting has been a subject of interest for many researchers; the essential market analyses can be integrated with historical stock market data to derive a set of features. It is crucial to select features with useful information about the specific aspect. In this article, we propose coefficient of variation (CV)-based feature selection for stock prediction. The unitless statistical method, CV, is widely used to obtain variability among data distributions. We calculate CV for each feature and integrate an existing method, $k$ -means algorithm, as well as proposed methods, median range and top- $M$ , to select a set of features with specific characteristics such as features belonging to the largest cluster, the defined range, and with the highest CV values, respectively. We apply the set of selected features to models such as backpropagation neural network (BPNN), long short-term memory (LSTM), gated recurrent unit (GRU), and convolutional neural network (CNN) for stock price and trend prediction. We demonstrate the applicability of our proposed approach using five of the existing feature selection methods, namely, correlation coefficient, Chi $^{2}$ , mutual information, principal component analysis, and variance threshold; comparison indicates remarkable performance enhancement using several accuracy-based, as well as error-based, metrics and the same is statistically supported using Wilcoxon signed-rank test.

View all citing articles on Scopus

¹: Tel.: +52 (81) 8158 2044.

View full text

ReviewGenetic algorithms and Darwinian approaches in financial applications: A survey

Highlights

Abstract

Introduction

Section snippets

Genetic algorithms and Darwinian approaches

Refining the scope of this work

Description of financial applications

Discussion and open problems

Conclusions

Acknowledgements

European Journal of Operational Research

Information Sciences

Expert Systems with Applications

Expert Systems with Applications

Expert Systems with Applications

Expert Systems with Applications

Information Sciences

Physica D: Nonlinear Phenomena

Information Processing Letters

Expert Systems with Applications

Expert Systems with Applications

Artificial Intelligence

Expert Systems with Applications

Applied Soft Computing

Decision Support Systems

Applied Soft Computing

Portfolio selection problem using generalized differential evolution 3

Applied Mathematical Sciences

Optimal execution of portfolio transactions

Journal of Risk

A comparative survey of artificial intelligence applications in finance: Artificial neural networks, expert system and hybrid intelligent systems

Neural Computing and Applications

A genetic type-2 fuzzy logic based system for financial applications modelling and prediction

On social security financial crisis

Journal of Population Economics

Agent-based computational macro-economics: A survey

Using extended classifier system to forecast S&P futures based on contrary sentiment indicators

A hybrid of adaptive genetic algorithm and pattern search for stock index optimized replicate

Generating trading rules on the stock markets with robust genetic network programming using variance of fitness values

Option pricing with genetic algorithms: A second report

Evolutionary models in cash management policies with multiple assets

Economic Modelling

An evolutionary morphological approach for financial time series forecasting

A fast elitist non-dominated sorting genetic algorithm for multi-objective optimization: NSGA-II

Lecture Notes in Computer Science

Comparative study of FOREX trading systems built with SVR+ GHSOM and genetic algorithms optimization of technical indicators

Forecasting time series by means of evolutionary algorithms

Evolutionary optimization of sparsely connected and time-lagged neural networks for time series forecasting

Applied Soft Computing

Gene expression programming: A new adaptive algorithm for solving problems

Complex Systems

Approximate and limited reasoning: Semantics, proof theory, expressivity and control

Journal of Logic And Computation

Coevolution and stable adjustments in the cobweb model

Journal of Evolutionary Economics

Multiobjective algorithms with resampling for portfolio optimization

Computing and Informatics

Self-adaptive MOEA feature selection for classification of bankruptcy prediction data

The Scientific World Journal

YACS: A new learning classifier system using anticipation

Soft Computing

Financial time series forecasting using agent based models in equity and FX markets

Building artificial economies: From aggregate data to experimental microstructure. A methodological survey

Genetic-fuzzy systems for financial decision making

Genetic algorithms for the traveling salesman problem

Asset portfolio optimization using support vector machines and real-coded genetic algorithm

Journal of Global Optimization

Applying dynamic training-subset selection methods using genetic programming for forecasting implied volatility

Computational Intelligence

The compact genetic algorithm

IEEE Transactions on Evolutionary Computation

Linkage learning via probabilistic modeling in the extended compact genetic algorithm (ECGA)

Optimization of the trading rule in foreign exchange using genetic algorithm

Review
Genetic algorithms and Darwinian approaches in financial applications: A survey