Financial forecasting through unsupervised clustering and neural networks

In this paper, we review our work on a time series forecasting methodology based on the combination of unsupervised clustering and artificial neural networks. To address noise and non-stationarity, a common approach is to combine a method for the partitioning of the input space into a number of subspaces with a local approximation scheme for each subspace. Unsupervised clustering algorithms have the desirable property of deciding on the number of partitions required to accurately segment the input space during the clustering process, thus relieving the user from making this ad hoc choice. Artificial neural networks, on the other hand, are powerful computational models that have proved their capabilities on numerous hard real-world problems. The time series that we consider are all daily spot foreign exchange rates of major currencies. The experimental results reported suggest that predictability varies across different regions of the input space, irrespective of clustering algorithm. In all cases, there are regions that are associated with a particularly high forecasting performance. Evaluating the performance of the proposed methodology with respect to its profit generating capability indicates that it compares favorably with that of two other established approaches. Moving from the task of one-step-ahead to multiple-step-ahead prediction, performance deteriorates rapidly.

[1]  Ah Chung Tsoi,et al.  Noisy Time Series Prediction using Recurrent Neural Networks and Grammatical Inference , 2001, Machine Learning.

[2]  Dimitris K. Tasoulis,et al.  Unsupervised distributed clustering , 2004, Parallel and Distributed Computing and Networks.

[3]  Dimitris K. Tasoulis,et al.  Improving the orthogonal range search k-windows algorithm , 2002, 14th IEEE International Conference on Tools with Artificial Intelligence, 2002. (ICTAI 2002). Proceedings..

[4]  Marie Cottrell,et al.  Some known facts about financial data , 2001, ESANN.

[5]  Ashok N. Srivastava,et al.  Nonlinear gated experts for time series: discovering regimes and avoiding overfitting , 1995, Int. J. Neural Syst..

[6]  Simon Haykin,et al.  Local Dynamic Modeling with SelfOrganizing Maps and Applications to Nonlinear System Identification and Control , 2001 .

[7]  Lijuan Cao,et al.  Support vector machines experts for time series forecasting , 2003, Neurocomputing.

[8]  Tony R. Martinez,et al.  Improved Heterogeneous Distance Functions , 1996, J. Artif. Intell. Res..

[9]  Maurice K. Wong,et al.  Algorithm AS136: A k-means clustering algorithm. , 1979 .

[10]  Dimitris K. Tasoulis,et al.  Computational Intelligence Methods for Financial Time Series Modeling , 2006, Int. J. Bifurc. Chaos.

[11]  Jingtao Yao,et al.  A case study on using neural networks to perform technical forecasting of forex , 2000, Neurocomputing.

[12]  Richard S. Sutton,et al.  Online Learning with Random Representations , 1993, ICML.

[13]  Andrew M. Fraser,et al.  Information and entropy in strange attractors , 1989, IEEE Trans. Inf. Theory.

[14]  C. Siriopoulos,et al.  Time series forecasting with a hybrid clustering scheme and pattern recognition , 2004, IEEE Transactions on Systems, Man, and Cybernetics - Part A: Systems and Humans.

[15]  J. Príncipe,et al.  Local dynamic modeling with self-organizing maps and applications to nonlinear system identification and control , 1998, Proc. IEEE.

[16]  Dimitris K. Tasoulis,et al.  Unsupervised clustering on dynamic databases , 2005, Pattern Recognit. Lett..

[17]  Halbert White,et al.  Connectionist nonparametric regression: Multilayer feedforward networks can learn arbitrary mappings , 1990, Neural Networks.

[18]  M. N. Vrahatis,et al.  Adaptive stepsize algorithms for on-line training of neural networks , 2001 .

[19]  Vassilis P. Plagianakos,et al.  Parallel evolutionary training algorithms for “hardware-friendly” neural networks , 2002, Natural Computing.

[20]  Rainer Storn,et al.  Differential Evolution – A Simple and Efficient Heuristic for global Optimization over Continuous Spaces , 1997, J. Glob. Optim..

[21]  Franklin Allen,et al.  Using genetic algorithms to find technical trading rules , 1999 .

[22]  Kenneth Rogoff,et al.  Was It Real? The Exchange Rate‐Interest Differential Relation over the Modern Floating‐Rate Period , 1988 .

[23]  Simon Haykin,et al.  Neural Networks: A Comprehensive Foundation , 1998 .

[24]  Steven Walczak,et al.  An Empirical Analysis of Data Requirements for Financial Forecasting with Neural Networks , 2001, J. Manag. Inf. Syst..

[25]  Kenneth S. Rogoff,et al.  Exchange rate models of the seventies. Do they fit out of sample , 1983 .

[26]  Michael N. Vrahatis,et al.  The New k-Windows Algorithm for Improving the k-Means Clustering Algorithm , 2002, J. Complex..

[27]  Eamonn J. Keogh,et al.  UCR Time Series Data Mining Archive , 1983 .

[28]  Irwin W. Sandberg,et al.  Uniform Approximation and Gamma Networks , 1997, Neural Networks.

[29]  Hans-Peter Kriegel,et al.  Density-Based Clustering in Spatial Databases: The Algorithm GDBSCAN and Its Applications , 1998, Data Mining and Knowledge Discovery.

[30]  Christian Igel,et al.  Improving the Rprop Learning Algorithm , 2000 .

[31]  H. Abarbanel,et al.  Determining embedding dimension for phase-space reconstruction using a geometrical construction. , 1992, Physical review. A, Atomic, molecular, and optical physics.

[32]  Irwin W. Sandberg,et al.  Uniform approximation of multidimensional myopic maps , 1997 .

[33]  Farmer,et al.  Predicting chaotic time series. , 1987, Physical review letters.

[34]  Maurice Clerc,et al.  The particle swarm - explosion, stability, and convergence in a multidimensional complex space , 2002, IEEE Trans. Evol. Comput..

[35]  C. Lee Giles,et al.  Learning Chaotic Attractors by Neural Networks , 2000, Neural Computation.

[36]  Andrew K. Rose,et al.  Empirical research on nominal exchange rates , 1995 .

[37]  Bernd Fritzke,et al.  A Growing Neural Gas Network Learns Topologies , 1994, NIPS.

[38]  Martin A. Riedmiller,et al.  A direct adaptive method for faster backpropagation learning: the RPROP algorithm , 1993, IEEE International Conference on Neural Networks.

[39]  Kenneth Rogoff,et al.  Empirical exchange rate models of the seventies , 1983 .

[40]  R. W. Dobbins,et al.  Computational intelligence PC tools , 1996 .

[41]  Allan Pinkus,et al.  Approximation theory of the MLP model in neural networks , 1999, Acta Numerica.

[42]  Ruy Luiz Milidiú,et al.  Time-series forecasting through wavelets transformation and a mixture of expert models , 1999, Neurocomputing.

[43]  Eric R. Ziegel,et al.  The Elements of Statistical Learning , 2003, Technometrics.

[44]  C. Diks,et al.  Detecting differences between delay vector distributions. , 1996, Physical review. E, Statistical physics, plasmas, fluids, and related interdisciplinary topics.

[45]  Apostolos-Paul Nicholas Refenes,et al.  Forecasting volatility with neural regression: A contribution to model adequacy , 2001, IEEE Trans. Neural Networks.

[46]  Martin Fodslette Meiller.  A Scaled Conjugate Gradient Algorithm for Fast Supervised Learning , 1993 .

[47]  Michael N. Vrahatis,et al.  Recent approaches to global optimization problems through Particle Swarm Optimization , 2002, Natural Computing.

[48]  F. Takens.  Detecting strange attractors in turbulence , 1981 .

[49]  James C. Bezdek,et al.  Pattern Recognition with Fuzzy Objective Function Algorithms , 1981, Advanced Applications in Pattern Recognition.

[50]  Dimitris K. Tasoulis,et al.  Parallelizing the Unsupervised k-Windows Clustering Algorithm , 2003, PPAM.

[51]  J. A. Hartigan,et al.  A k-means clustering algorithm , 1979 .

[52]  R. Storn,et al.  Differential Evolution - A simple and efficient adaptive scheme for global optimization over continuous spaces , 2004 .

[53]  Dimitris K. Tasoulis,et al.  Time Series Forecasting Methodology for Multiple-Step-Ahead Prediction , 2005, Computational Intelligence.

[54]  Hans-Peter Kriegel,et al.  A Density-Based Algorithm for Discovering Clusters in Large Spatial Databases with Noise , 1996, KDD.