An experimental study of Multi-Objective Evolutionary Algorithms for balancing interpretability and accuracy in fuzzy rulebase classifiers for financial prediction

This paper examines the advantages of simple models over more complex ones for financial prediction. This premise is examined using a genetic fuzzy framework. The interpretability of fuzzy systems is oftentimes put forward as a unique advantageous feature, sometimes to justify effort associated with using fuzzy classifiers instead of alternatives that can be more readily implemented using existing tools. Here we investigate if model interpretability can provide further benefits by realizing useful properties in computationally intelligent systems for financial modeling. We test an approach for learning momentum based strategies that predict price movements of the Bombay Stock Exchange (BSE). The paper contributes an experimental evaluation of the relationship between the predictive capability and interpretability of fuzzy rule based systems obtained using Multi-Objective Evolutionary Algorithms (MOEA.)

[1]  Zbigniew Michalewicz,et al.  A Computational Intelligence Portfolio Construction System for Equity Market Trading , 2007, 2007 IEEE Congress on Evolutionary Computation.

[2]  Kim-Fung Man,et al.  Multi-objective hierarchical genetic algorithm for interpretable fuzzy rule-based knowledge extraction , 2005, Fuzzy Sets Syst..

[3]  J. van Leeuwen,et al.  Evolutionary Multi-Criterion Optimization , 2003, Lecture Notes in Computer Science.

[4]  David Corne,et al.  The Pareto archived evolution strategy: a new baseline algorithm for Pareto multiobjective optimisation , 1999, Proceedings of the 1999 Congress on Evolutionary Computation-CEC99 (Cat. No. 99TH8406).

[5]  Hisao Ishibuchi,et al.  Interactive Fuzzy Modeling by Evolutionary Multiobjective Optimization with User Preference , 2009, IFSA/EUSFLAT Conf..

[6]  Kenneth A. De Jong,et al.  Cooperative Coevolution: An Architecture for Evolving Coadapted Subcomponents , 2000, Evolutionary Computation.

[7]  Francisco Herrera,et al.  Ten years of genetic fuzzy systems: current framework and new trends , 2004, Fuzzy Sets Syst..

[8]  Enrique Alba,et al.  MOCell: A cellular genetic algorithm for multiobjective optimization , 2009, Int. J. Intell. Syst..

[9]  Hisao Ishibuchi,et al.  Analysis of interpretability-accuracy tradeoff of fuzzy systems by multiobjective fuzzy genetics-based machine learning , 2007, Int. J. Approx. Reason..

[10]  Jan Paredis,et al.  Coevolutionary computation , 1995 .

[11]  Martin J. Oates,et al.  The Pareto Envelope-Based Selection Algorithm for Multi-objective Optimisation , 2000, PPSN.

[12]  Enrique Alba,et al.  AbYSS: Adapting Scatter Search to Multiobjective Optimization , 2008, IEEE Transactions on Evolutionary Computation.

[13]  Jose M. Benjorge,et al.  Fuzzy Control of HVAC Systems Optimized by Genetic Algorithms , 2003 .

[14]  Anna Maria Fanelli,et al.  Interpretability Assessment of Fuzzy Rule-Based Classifiers , 2009, WILF.

[15]  H. Ishibuchi,et al.  Minimizing the fuzzy rule base and maximizing its performance by a multiobjective genetic algorithm , 1997, Proceedings of 6th International Fuzzy Systems Conference.

[16]  E. Fama,et al.  Multifactor Explanations of Asset Pricing Anomalies , 1996 .

[17]  Marco Laumanns,et al.  SPEA2: Improving the strength pareto evolutionary algorithm , 2001 .

[18]  Zbigniew Michalewicz,et al.  Learning Fuzzy Rules with Evolutionary Algorithms - An Analytic Approach , 2008, PPSN.

[19]  John Yen,et al.  Improving the interpretability of TSK fuzzy models by combining global learning and local learning , 1998, IEEE Trans. Fuzzy Syst..

[20]  Zbigniew Michalewicz,et al.  GENOCOP: a genetic algorithm for numerical optimization problems with linear constraints , 1996, CACM.

[21]  R. K. Ursem.  Multi-objective Optimization using Evolutionary Algorithms , 2009 .

[22]  Nicola Beume,et al.  SMS-EMOA: Multiobjective selection based on dominated hypervolume , 2007, Eur. J. Oper. Res..

[23]  Kalyanmoy Deb,et al.  A fast and elitist multiobjective genetic algorithm: NSGA-II , 2002, IEEE Trans. Evol. Comput..

[24]  Hisao Ishibuchi,et al.  Finding Simple Fuzzy Classification Systems with High Interpretability Through Multiobjective Rule Selection , 2006, KES.

[25]  Jennifer Conrad,et al.  Profitability of Momentum Strategies: An Evaluation of Alternative Explanations , 2001 .

[26]  Hisao Ishibuchi,et al.  Incorporation of decision maker's preference into evolutionary multiobjective optimization algorithms , 2006, GECCO '06.

[27]  Martin J. Oates,et al.  PESA-II: region-based selection in evolutionary multiobjective optimization , 2001 .

[28]  Enrique Alba,et al.  The jMetal framework for multi-objective optimization: Design and architecture , 2010, IEEE Congress on Evolutionary Computation.

[29]  Hamidreza Eskandari,et al.  FastPGA: A Dynamic Population Sizing Approach for Solving Expensive Multiobjective Optimization Problems , 2006, EMO.

[30]  Peter J. Fleming,et al.  Evolutionary many-objective optimisation: an exploratory analysis , 2003, The 2003 Congress on Evolutionary Computation, 2003. CEC '03..

[31]  Kalyanmoy Deb,et al.  Multi-objective optimization using evolutionary algorithms , 2001, Wiley-Interscience series in systems and optimization.

[32]  Francisco Herrera,et al.  A learning process for fuzzy control rules using genetic algorithms , 1998, Fuzzy Sets Syst..

[33]  Zbigniew Michalewicz,et al.  Computational Intelligence for Evolving Trading Rules , 2009, IEEE Transactions on Evolutionary Computation.

[34]  Ludmila I. Kuncheva,et al.  How good are fuzzy If-Then classifiers? , 2000, IEEE Trans. Syst. Man Cybern. Part B.

[35]  María José del Jesús,et al.  Genetic tuning of fuzzy rule deep structures preserving interpretability and its interaction with fuzzy rule set reduction , 2005, IEEE Transactions on Fuzzy Systems.

[36]  Chung-Ming Kuan,et al.  Reexamining the Profitability of Technical Analysis with Data Snooping Checks , 2005 .