KOSPI Time Series Analysis Using Neural Network with Weighted Fuzzy Membership Functions

Fuzzy neural networks have been successfully applied to generate predictive rules for stock forecasting. This paper presents a methodology to forecast the daily Korea composite stock price index (KOSPI) by extracting fuzzy rules based on the neural network with weighted fuzzy membership functions (NEWFM) and the minimized number of input features using the distributed non-overlap area measurement method. NEWFM supports the KOSPI time series analysis based on the defuzzyfication of weighted average method which is the fuzzy model suggested by Takagi and Sugeno. NEWFM classifies upward and downward cases of next day's KOSPI using the recent 32 days of CPPn,m (Current Price Position of day n : a percentage of the difference between the price of day n and the moving average of the past m days from day n-1) of KOSPI. In this paper, the Haar wavelet function is used as a mother wavelet. The most important five input features among CPPn,m and 38 numbers of wavelet transformed coefficients produced by the recent 32 days of CPPn,m are selected by the non-overlap area distribution measurement method. For the data sets, from 1991 to 1998, the proposed method shows that the average of accuracy rate is 67.62%.

[1]  Johan A. K. Suykens,et al.  Financial time series prediction using least squares support vector machines within the evidence framework , 2001, IEEE Trans. Neural Networks.

[2]  Patrick K. Simpson,et al.  Fuzzy min-max neural networks. I. Classification , 1992, IEEE Trans. Neural Networks.

[3]  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.

[4]  Stéphane Mallat,et al.  Zero-crossings of a wavelet transform , 1991, IEEE Trans. Inf. Theory.

[5]  Hisao Ishibuchi,et al.  Voting in fuzzy rule-based systems for pattern classification problems , 1999, Fuzzy Sets Syst..

[6]  Jyh-Shing Roger Jang,et al.  ANFIS: adaptive-network-based fuzzy inference system , 1993, IEEE Trans. Syst. Man Cybern..

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

[8]  Jeen-Shing Wang,et al.  Self-adaptive neuro-fuzzy inference systems for classification applications , 2002, IEEE Trans. Fuzzy Syst..

[9]  Dianhui Wang,et al.  A neuro-fuzzy approach for diagnosis of antibody deficiency syndrome , 2006, Neurocomputing.

[10]  Magne Setnes,et al.  GA-fuzzy modeling and classification: complexity and performance , 2000, IEEE Trans. Fuzzy Syst..

[11]  Michio Sugeno,et al.  Fuzzy identification of systems and its applications to modeling and control , 1985, IEEE Transactions on Systems, Man, and Cybernetics.

[12]  P. K. Simpson.  Fuzzy Min-Max Neural Networks-Part 1 : Classification , 1992 .

[13]  Xiaohua Wang,et al.  Stock market prediction using neural networks: Does trading volume help in short-term prediction? , 2003, Proceedings of the International Joint Conference on Neural Networks, 2003..

[14]  Joon Shik Lim.  Finding Fuzzy Rules for IRIS by Neural Network with Weighted Fuzzy Membership Function , 2004, Int. J. Fuzzy Log. Intell. Syst..

[15]  Kyoung-jae Kim,et al.  Financial time series forecasting using support vector machines , 2003, Neurocomputing.

[16]  Rudolf Kruse,et al.  A neuro-fuzzy method to learn fuzzy classification rules from data , 1997, Fuzzy Sets Syst..

[17]  Ho Joon Kim,et al.  Feature Selection for Specific Antibody Deficiency Syndrome by Neural Network with Weighted Fuzzy Membership Functions , 2005, FSKD.