CARTMAP: a neural network method for automated feature selection in financial time series forecasting

In the past two decades, there has been much interest in applying neural networks to financial time series forecasting. Yet, there has been relatively little attention paid to selecting the input features for training these networks. This paper presents a novel CARTMAP neural network based on Adaptive Resonance Theory that incorporates automatic, intuitive, transparent, and parsimonious feature selection with fast learning. On average, over three separate 4-year simulations spanning 2004–2009 of Dow Jones Industrial Average stocks, CARTMAP outperformed related and classical alternatives. The alternatives were an industry standard random walk, a regression model, a general purpose ARTMAP, and ARTMAP with stepwise feature selection. This paper also discusses why the novel feature selection scheme outperforms the alternatives and how it can represent a step toward more transparency in financial modeling.

[1]  Andrew W. Lo,et al.  Bubble, Rubble, Finance in Trouble? , 2002 .

[2]  Yevgeniy V. Bodyanskiy,et al.  Neural network approach to forecasting of quasiperiodic financial time series , 2006, Eur. J. Oper. Res..

[3]  Gail A. Carpenter,et al.  ARTMAP neural networks for information fusion and data mining: map production and target recognition methodologies , 2003, Neural Networks.

[4]  Jonathan D. Cryer,et al.  Time Series Analysis , 1986 .

[5]  Mitsuo Gen,et al.  Recurrent neural network for dynamic portfolio selection , 2006, Appl. Math. Comput..

[6]  Stephen Grossberg,et al.  Fuzzy ART: Fast stable learning and categorization of analog patterns by an adaptive resonance system , 1991, Neural Networks.

[7]  W. Sharpe.  The Sharpe Ratio , 1994 .

[8]  Gail A. Carpenter,et al.  Default ARTMAP 2 , 2007, 2007 International Joint Conference on Neural Networks.

[9]  Hong Wang,et al.  Predicting stock index increments by neural networks: The role of trading volume under different horizons , 2008, Expert Syst. Appl..

[10]  Kin Keung Lai,et al.  Neural network-based mean-variance-skewness model for portfolio selection , 2008, Comput. Oper. Res..

[11]  David West,et al.  Neural network credit scoring models , 2000, Comput. Oper. Res..

[12]  Yannis Manolopoulos,et al.  Data Mining techniques for the detection of fraudulent financial statements , 2007, Expert Syst. Appl..

[13]  D L Woods,et al.  Electrophysiologic evidence of increased distractibility after dorsolateral prefrontal lesions , 1986, Neurology.

[14]  P. H. Miller,et al.  Strategy utilization deficiencies in children: when, where, and why. , 1994, Advances in child development and behavior.

[15]  Michael F. Bunting,et al.  The cocktail party phenomenon revisited: The importance of working memory capacity , 2001, Psychonomic bulletin & review.

[16]  Milton S. Boyd,et al.  Designing a neural network for forecasting financial and economic time series , 1996, Neurocomputing.

[17]  Soushan Wu,et al.  Comparison of support-vector machines and back propagation neural networks in forecasting the six major Asian stock markets , 2006 .

[18]  David G. Stork,et al.  Pattern Classification , 1973 .

[19]  Massimiliano Versace,et al.  Predicting the exchange traded fund DIA with a combination of genetic algorithms and neural networks , 2004, Expert Syst. Appl..

[20]  B. Malkiel.  A Random Walk Down Wall Street , 1973 .

[21]  W. Sharpe.  CAPITAL ASSET PRICES: A THEORY OF MARKET EQUILIBRIUM UNDER CONDITIONS OF RISK* , 1964 .

[22]  R. Engle,et al.  The role of prefrontal cortex in working-memory capacity, executive attention, and general fluid intelligence: An individual-differences perspective , 2002, Psychonomic bulletin & review.

[23]  E. Pasour.  The efficient-markets hypothesis and entrepreneurship , 1989 .

[24]  J. Murphy.  Technical Analysis of the Financial Markets , 1999 .

[25]  E. Fama.  EFFICIENT CAPITAL MARKETS: A REVIEW OF THEORY AND EMPIRICAL WORK* , 1970 .

[26]  M. Posner,et al.  The development of orienting to locations and objects , 1998 .

[27]  Ingoo Han,et al.  Genetic algorithms approach to feature discretization in artificial neural networks for the prediction of stock price index , 2000 .

[28]  David Enke,et al.  The adaptive selection of financial and economic variables for use with artificial neural networks , 2004, Neurocomputing.

[29]  C. Kirkpatrick,et al.  Technical Analysis: The Complete Resource for Financial Market Technicians , 2006 .

[30]  Hubert M. Blalock,et al.  Correlated Independent Variables: The Problem of Multicollinearity , 1963 .

[31]  Keming Wang.  Neural Network Approach to Vibration Feature Selection and Multiple Fault Detection for Mechanical Systems , 2006, First International Conference on Innovative Computing, Information and Control - Volume I (ICICIC'06).

[32]  Byung Ro Moon,et al.  A Hybrid Neurogenetic Approach for Stock Forecasting , 2007, IEEE Transactions on Neural Networks.

[33]  J. Quiggin.  THE EFFICIENT MARKETS HYPOTHESIS , 2010 .

[34]  G. Edwards,et al.  The Theory of Investment Value. , 1939 .

[35]  Gerald M. Loeb,et al.  The Battle for Investment Survival , 1965 .

[36]  Radford M. Neal.  Pattern Recognition and Machine Learning , 2007, Technometrics.

[37]  Andrew Chi-Sing Leung,et al.  Combined learning and pruning for recurrent radial basis function networks based on recursive least square algorithms , 2005, Neural Computing & Applications.

[38]  Charles Wong,et al.  Context Sensitivity with Neural Networks in Financial Decision Processes , 2011 .

[39]  Tianzi Jiang,et al.  A combinational feature selection and ensemble neural network method for classification of gene expression data , 2004, BMC Bioinformatics.

[40]  Jaeho Jang,et al.  A Comparison of Neural Network, Statistical Methods, and Variable Choice for Life Insurers' Financial Distress Prediction , 2006 .

[41]  Stephen Grossberg,et al.  ARTMAP: supervised real-time learning and classification of nonstationary data by a self-organizing neural network , 1991, [1991 Proceedings] IEEE Conference on Neural Networks for Ocean Engineering.

[42]  Jack D. Schwager.  The New Market Wizards: Conversations with America's Top Traders , 1992 .

[43]  Itzhak Gilboa,et al.  Fact-Free Learning , 2004 .

[44]  J. J. Higgins.  Introduction to Modern Nonparametric Statistics , 2003 .

[45]  J. Murphy.  Technical Analysis of the Futures Markets: A Comprehensive Guide to Trading Methods and Applications , 1986 .

[46]  Kyoung-jae Kim.  Artificial neural networks with evolutionary instance selection for financial forecasting , 2006, Expert Syst. Appl..

[47]  H. Markowitz.  The Utility of Wealth , 1952, Journal of Political Economy.

[48]  Stephen Grossberg,et al.  A massively parallel architecture for a self-organizing neural pattern recognition machine , 1988, Comput. Vis. Graph. Image Process..

[49]  J. Lintner.  THE VALUATION OF RISK ASSETS AND THE SELECTION OF RISKY INVESTMENTS IN STOCK PORTFOLIOS AND CAPITAL BUDGETS , 1965 .

[50]  Steven Thorley,et al.  Mining Fool's Gold , 1999 .

[51]  M. Alibali,et al.  Children's Thinking , 1986 .

[52]  Kuldeep Kumar,et al.  Artificial neural network vs linear discriminant analysis in credit ratings forecast: A comparative study of prediction performances , 2006 .

[53]  R. Knight,et al.  Contribution of Human Prefrontal Cortex to Delay Performance , 1998, Journal of Cognitive Neuroscience.

[54]  L. Blume,et al.  The New Palgrave Dictionary of Economics, 2nd edition , 2008 .

[55]  Ian Witten,et al.  Data Mining , 2000 .