臺灣博碩士論文加值系統

股票投資是一種極具挑戰性且令人興奮的投資理財活動，股票指數波動瞬息萬變，盈虧結果取決於正確抉擇。因此，一個可靠而精準之預測模型是股票投資者至為重要的決策輔助參考工具。時間序列預測模型在趨勢分析與決策過程中扮演著極重要的決策輔助角色，許多研究者針對時間序列的預測問題提出了各類型的預測模型。傳統時間序列模型只能解決數值型態的問題，對語意型歷史資料卻無法處理，而模糊時間序列就是一種針對解決語意型資料而發展出來的預測模型。早期的模糊時間序列研究，在論域上之定義，是以兩個正整數值作為論域上下界限之延伸，然後再依語意數，將論域做等距區間之切割，此種切割方法忽略資料集之分佈特性，且其將各模糊邏輯關係視為均等重要。本論文將提出一種基於累積機率分配方法與粗集方法之整合性模糊時間序列模型，以解決上述傳統模糊時間序列模型的問題，該整合性模糊時間序列模型將探討資料分佈性與模糊邏輯關係出現頻率在股票指數預測之影響及其重要性。本論文提出三種不同的整合性模型: (1)整合累積機率分配方法、粗集方法與模糊時間序列在股票指數之預測(稱為模型A)，(2)整合粗集方法與模糊時間序列在股票指數之預測(稱為模型B)，及(3)整合累積機率分配方法與模糊時間序列在股票指數之預測(稱為模型C)。本論文並使用三個股票指數資料集，即台灣股票加權指數(TAIEX)、紐約綜合股票指數(NYSE)及日經指數(NIKKEI)，以驗證研究模型應用在多階股票指數預測之正確性，以及使用台灣股票加權指數資料集(TAIEX)來驗證研究模型於多技術指標股票指數預測之正確性。本論文採用平均誤差平方和之平方根(root mean square error)及平均絕對百分誤差(mean absolute percent error)作為正確性比較準則，並將研究模型預測結果與Chen’s (1996), Yu’s (2005) 及 Huarng (2006)等模型作一比較。

Stock investment is one of most exciting and challenging monetary activity. The market climates are dramatically in minute and gain-loss decides in twinkle decision, thus an accurate forecasting tool is crucial. Many researchers have presented different forecasting models to deal with forecasting problems on time series. Time series forecasting model plays an important role in trend analysis and decision making. Traditional time series models can only solve numerical type problem, but fail to treat the problems with linguistic historical data. To deal with these kinds of problems, an alternative forecasting model such as fuzzy time series is needed. However, previous studies of fuzzy time series often determine the universe of discourse with a simply chosen positive integer to extend both boundaries, partition the universe of discourse into equal-length linguistic intervals, ignore the distribution characteristic of dataset and treat each fuzzy relationship as being of equal importance. To reconcile drawbacks of those conventional models, this dissertation proposed an integrated model of fuzzy time-series for forecasting stock index based on cumulative probability distribution approach (CPDA) and rough set approach, which will consider the distribution characteristics of data and recurrent fuzzy relationships. This dissertation proposes a set of refined cause-and-effect models which include three different integrated models: (1) fuzzy time-series for forecasting stock index based on CPDA and rough set approach (model A), (2) fuzzy time-series for forecasting stock index based on rough sets approach (model B), and (3) fuzzy time-series for forecasting stock index based on CPDA (model C). The effectiveness of the proposed models are verified by multi orders forecasting using three stock index databases (TAIEX, NYSE and NIKKEI)and multi attributes forecasting using TAIEX database. The forecasting performance are compared with Chen’s (1996), Yu’s (2005) and Huarng (2006) models with root of mean square error (RMSE) and mean absolute percent error (MAPE).

中 文 摘 要 VI
ABSTRACT VIII
誌 謝 X
1. INTRODUCTION 1
1.1 BACKGROUND 1
1.2 MOTIVATION AND RESEARCH OBJECTIVE 3
1.3 RESEARCH FRAMEWORK AND ORGANIZATION OF THIS DISSERTATION 6
2. LITERATURE REVIEW 9
2.1 FUZZY SETS THEORY 9
2.1.1 Fuzzy numbers 11
2.1.2 Linguistic variable and linguistic value 13
2.1.3 Defuzzification 15
2.2 PROBABILITY DISTRIBUTIONS AND CUMULATIVE DISTRIBUTION FUNCTIONS 16
2.3 FUZZY TIME SERIES 18
2.4 ROUGH SETS THEORY 20
2.4.1 Information table 24
2.4.2 Reduction and dependency of attributes 25
2.4.3 Decision rules 25
2.4.4 Decision support using decision rules 27
2.5 ADAPTIVE EXPECTATION MODEL 27
2.6 STOCK PRICE FORECASTING 28
2.7 MARKET ANALYSIS 29
3. RESEARCH FRAMEWORKS AND PROPOSED ALGORITHMS 34
3.1 RESEARCH FRAMEWORKS 34
3.2. THE ALGORITHMS OF THESE THREE PROPOSED MODELS 40
3.2.1 Fuzzy time-series for forecasting stock index based on CPDA and rough set approach (model A) 41
3.2.2 Fuzzy time-series for forecasting stock index based on rough set approach (model B) 62
3.2.3 Fuzzy time-series for forecasting stock index based on CPDA (model C) 66
4. VERIFICATIONS OF PROPOSED MODELS ON MULTI ORDERS FORECASTING 72
4.1 PERFORMANCE EVALUATION USING THREE DIFFERENT LINGUISTIC VALUES 74
4.2 PERFORMANCE VERIFICATION ON ONE-ORDER FORECASTING 75
4.2.1 Empirical analysis for TAIEX database. 76
4.2.2 Empirical analysis for NYSE database. 78
4.2.3 Empirical analysis for NYSE database. 80
4.3 PERFORMANCE VERIFICATION ON HIGH-ORDER FORECASTING 82
4.3.1 Empirical analysis for TAIEX database. 83
4.3.2 Empirical analysis for NYSE database. 87
4.3.3. Empirical analysis for NIKKEI database. 91
5. VERIFICATIONS OF PROPOSED MODELS ON MULTI ATTRIBUTES FORECASTING 96
5.1 PRODUCE THE TEN TECHNICAL INDICATORS 98
5.2 THE PROCEDURE OF MULTI ATTRIBUTES FORECASTING 99
5.2 EMPIRICAL RESULTS 102
6. FINDINGS 103
1 FINDING ON ONE-ORDER FORECASTING PERFORMANCE 103
2 FINDING ON HIGH-ORDER FORECASTING PERFORMANCE 106
3 FINDING ON MULTI ATTRIBUTES FORECASTING PERFORMANCE 112
7. CONCLUSIONS AND FUTURE WORKS 113
REFERENCE 118
APPENDIX I PSEUDO-CODE ALGORITHM FOR RATIONAL APPROXIMATION 124
APPENDIX II PSEUDO-CODE ALGORITHM FOR THE LEM2 PROCEDURE 126
APPENDIX III THE ECONOMICAL MEANING AND FORMULA OF TECHNICAL INDICATOR 127
PUBLICATION LIST 129

[1].Acklam, P. J. (2004). An algorithm for computing the inverse normal cumulative distribution function.
[2].Bensignor, R. (2001). New thinking in technical analysis: trading models from the masters. Milton: wrightbooks.
[3].Berkan, R. C., & Trubatch, S. L. (1997). Fuzzy systems design principles - building fuzzy If-Then rule bases. New York: IEEE Press.
[4].Beynon, M. J., & Peel, M. J. (2001). Variable precision rough set theory and data discretisation: an application to corporate failure prediction. Omega - The International Journal of Management Science, 29, 561-576.
[5].Bojadziev, G., & Bojadziev, M. (1997). Fuzzy logic for business, finance, and management. River Edge, NJ: World Scientific.
[6].Brane, M., & Rutherford, D. A. (1978). Fuzzy relations in a control setting. Kybenetes, 7(3), 185-188.
[7].Cao, L. J., & Tay, F. E. H. (2003). Support Vector Machine with Adaptive Parameters in Financial Time Series Forecasting. IEEE Trans. Neural Networks, 14(6), 1506-1518.
[8].Chen, S. J., & Chen, S. M. (2005). Aggregating fuzzy opinions in the heterogeneous group decision-making environment,. Cybernetics and Systems: An International Journal, 36(3), 309-338.
[9].Chen, S. M. (1996). Forecasting enrollments based on fuzzy time series. Fuzzy Sets and Systems, 81, 311-319.
[10].Chen, S. M. (2002). Forecasting enrollments based on high-order fuzzy time series. Cybernetics and Systems, 33, 1-16.
[11].Chen, S. M., & Chung, N. Y. (2006). Forecasting enrollments using high-order fuzzy time series and genetic algorithms. International Journal of Intelligent Systems, 21 485-501.
[12].Chen, S. M., & Hsu, C. C. (2004). A new method to forecast enrollments using fuzzy time series. Applied Science and Eng, 2, 234-244.
[13].Chen, T.-L., Cheng, C.-H., & Teoh, H. J. (2007). Fuzzy time-series based on Fibonacci sequence for stock price forecasting. Physica A, 380, 377-390.
[14].Chen, T. L., Cheng, C. H., & Teoh, H. J. (2007). High-order fuzzy time-series based on multi-period adaptation model for forecasting stock markets. Physica A: Statistical Mechanics and its Applications.
[15].Cheng, C. H., Chen, T. L., & Chiang, C. H. (2006). Trend-weighted fuzzy time-series model for TAIEX forecasting. Lect. Notes Comput. Sci, 4234, 469-477.
[16].Cheng, C. H., Teoh, H. J., & Chen, T. L. (2007a). Forecasting Stock Price Index Using Fuzzy Time-Series Based on Rough Set. Paper presented at the FSKD 2007, Hainan.
[17].Cheng, C. H., Teoh, H. J., & Chen, T. L. (2007b). High-Order Fuzzy Time Series Based on Rough Set for Forecasting Taiex. Paper presented at the Machine Learning and Cybernetics, Hong Kong.
[18].Cheng, C. H., Teoh, H. J., & Li, C. H. (2007). Fuzzy Weighted-Transitional Matrix for Forecasting Time Series Data. Paper presented at the Machine Learning and Cybernetics Hong Kong.
[19].Devore, J. L. (2004). Probability and Statistics for Engineering and the Sciences (sixth ed.). Belmont: Duxbury.
[20].Dhar, V., & Chou, D. (2001). A Comparison of Nonlinear Methods for Predicting Earnings Surprises and Returns. IEEE Trans. Neural Networks, 12(4), 907-921.
[21].Dubois, D., & Prade, H. (1978). Operations on fuzzy numbers. Internat. J. Systems Science, 9 (6), 613-626.
[22].Dubois, D., Prade, H., & . (1988). Possibility Theory: An Approach to Computerized Processing of Uncertainty. New York: Plenum Press.
[23].Grzymala-Busse, J. (1992). LERS—a system for learning from examples based on rough sets. Dordrecht: Kluwer Academic Publishers.
[24].Grzymala-Busse, J. W. (1997). A new version of the rule induction system LERS. Fundamenta Informaticae, 31, 27-39.
[25].Huarng, K. (2001). Heuristic models of fuzzy time series for forecasting Fuzzy Sets and Systems, 123, 137-154.
[26].Huarng, K., & Yu, H.-k. (2003). An N-th order heuristic fuzzy time series model for TAIEX forecasting. internat. J. Fuzzy Syztems, 5(4), 247-253.
[27].Huarng, K. H. (2001). Effective lengths of intervals to improve forecasting in fuzzy time-series. Fuzzy Sets and Systems, 123, 387-395.
[28].Hwang, J. R., Chen, S.-M., & Lee, C.-H. (1998). Handling forecasting problems using fuzzy time series. Fuzzy Sets and Systems, 100, 217-228.
[29].Kahn, M. N. (2006). Technical Analysis Plain and Simple - Charting the markets in your language (second ed.). New Jersey: Prentice Hall.
[30].Kandel, M., & Friedman. (1998). Defuzzification using most optical values. IE/EE Trans on Sytems, man & Cybernetics part B, 28(6), 901-906.
[31].Kaufmann, A., & Gupta, M. M. (1988). Fuzzy mathematical models in engineering and management science. Amsterdam: North-Holland
[32].Kendall, S. M., & Ord, K. (1990). Time series (3rd ed.). New York: Oxford university press.
[33].Kim, M.-J., Min, S.-H., & Han, I. (2006). An evolutionary approach to the combination of multiple classifiers to predict a stock price index. Expert System with applications, 31(2), 241-247.
[34].Kmenta, J. (1986). Elements of Econometrics: MacMillan.
[35].Kumar, P., Krishna, P. R., Bapi, R. S., & De, S. K. (2007). Rough clustering of sequential data. Data & Knowledge Engineering, 63, 183-199.
[36].Kuo, R. J., Chen, C. H., & Hwang, Y. C. (2001). An intelligence stock trading decision support system through integration of genetic algorithm based fuzzy neural network and artificial neural network. Fuzzy Sets and Systems, 118, 21-45.
[37].Lin, H. L., & Chen, C. J. (2004). Basic statistics concepts and applications (Vol. 1). Taipei: Yeh yeh book gallery.
[38].Lin, S. Y. (2000). Forecasting Taiwan Stock Index Future-An Application of Fuzzy Time series. Chaoyang University of Technology, Taichung County.
[39].McLaughlin, M. P. (2001, 2007/8/11). A Compendium of Common Probability Distributions. Retrieved 8/13, 2007, from http://en.wikipedia.org/wiki/Probability_distribution
[40].Mehmed, K. (2003). Data Mining: Concepts, Models, Methods, and Algorithm: John Wiley & Sons.
[41].Miller, G. A. (1994). The magical number seven, plus or minus two: Some limits on our capacity of processing information. The Psychological Review, 101, 343-352.
[42].Pantazopoulos, K. N., Tsoukalas, L. H., Bourbakis, N. G., Brun, M. J., & Houstis, E. N. (1998). Financial Prediction and Trading Strategies Using Neurofuzzy Approaches IEEE Trans. Systems, Man, and Cybernetics and Systems, 28(4), 520-531.
[43].Parmar, D., Wu, T., & Blackhurst, J. (2007). MMR: An algorithm for clustering categorical data using Rough Set Theory. Data & Knowledge Engineering, 63, 879-893.
[44].Pawlak, Z. (1982). Rough sets. Internat.J.Comput.Inform.Sci, 5 341-356.
[45].Pawlak, Z. (1991). Rough sets : theoretical aspects of reasoning about data. Dordrecht ; ; Boston: Kluwer Academic Publishers.
[46].Pawlak, Z. (1997). Rough set approach to knowledge-based decision support. European Journal of Operational Research 99, 48-57.
[47].Pawlak, Z. (2002). Rough sets and intelligent data analysis Information Sciences, 147, 1-12.
[48].Pawlak, Z., Grzymala-Busse, J., Slowinski, R., & Ziarko, W. (1995). Rough sets. . Association for Computing Machinery: Communications of the ACM, 38(11), 89-96.
[49].Pawlak, Z., & Skoworn, A. (2007). Rudiments of rough sets. Information Science, 177, 3-27.
[50].Pawlak, Z., & Slowinski, R. (1994). Rough set approach to multiattribute decision analysis. European Journal of Operational Research, 72, 443-459.
[51].Refenes, A. P. N., & Holt, W. T. (2001). Forecasting Volatility with Neural Regression: A Contribution to Model Adequacy. IEEE Trans. Neural Networks, 12(4), 850-864.
[52].Reuters. (1999). An Introduction to Technical Analysis. singapore: wiley.
[53].Ross, T. J. (2004). Fuzzy logic with engineering applications (2nd ed.). Hoboken, NJ: John Wiley.
[54].Saad, E. W., Prokhorov, D. V., & Wunsch, D. C. I. (1998). Comparative Study of Stock Trend Prediction Using Time Delay, Recurrent and Probabilistic Neural Networks. IEEE Trans. Neural Networks, 9(6), 1456-1470.
[55].Shieh, J. C. P. (1998.). Contemporary Investments-Analysis And Management. Taipei, Taiwan:: Best-Wise.
[56].Slowinski, R. (1992). Handbook of Applications and Advances of the Rough Sets Theory. Dordrecht: Kluwer Academic Publishers.
[57].Slowinski, R. (1995). Rough set approach to decision analysis. AI Expert Magazine, 10, 18-25.
[58].Slowinski, R., & Stefanowski, J. (1994). New Approaches in Classification and Data Analysis. Berlin: Springer.
[59].Song, Q., & Chissom, B. S. (1993a). Forecasting enrollments with fuzzy time series - Part Ⅰ. Fuzzy Sets and Systems, 54, 1-10.
[60].Song, Q., & Chissom, B. S. (1993b). Fuzzy time series and its models. Fuzzy Sets and Systems, 54(3), 269-277.
[61].Song, Q., & Chissom, B. S. (1994). Forecasting enrollments with fuzzy time-series - Part Ⅱ. Fuzzy Sets and Systems, 62, 1-8.
[62].Tanaka, H. (1987). Fuzzy data analysis by possibility linear models. Fuzzy Sets and Systems, 24, 363-375.
[63].Tanaka, H., & Ishibuchi, H. (1992). Possibility regression analysis based on linear programming. Physica-Verlag, J. Kacprzyk, M. Fedrizzi, 47–60.
[64].Tanaka, H., Uejima, S., & Asai, K. (1982). Linear regression analysis with fuzzy model. IEEE Trans. Syst. Man, Cybernetics and Systems, 12(6), 903-907.
[65].Tanaka, K., & Wang, H. O. (2001). Fuzzy control systems design and analysis : a linear matrix inequality approach. New York: Wiley.
[66].Teoh, H. J., Chen, T. L., & Cheng, C. H. (2007). Frequency-Weighted Fuzzy Time-Series Based on Fibonacci Sequence for TAIEX Forecasting. Paper presented at the PAKDD 2007 Workshops, Nanjing.
[67].Tsai, C. C., & Wu, S. J. (1999). A Study for Second-order Modeling of Fuzzy Time Series. Paper presented at the Proc. IEEE Fuzzy System Conf.
[68].Tseng, F.-M., & Tzeng, G.-H. (2002). Fuzzy seasonal ARIMA model for forecasting. Fuzzy Sets and Systems, 126, 367-376.
[69].Tseng, F. M., Tzeng, G. H., Yu, H. C., & Yuan, B. J. C. (2000). Fuzzy ARIMA model for forecasting the foreign exchange market. Fuzzy Sets and Systems, 118, 9-19.
[70].Wakulicz-Deja, A., & Paszek, P. (1997). International Diagnose progressive encephalopathy applying the rough set approach. Journal of Medical Informatics, 46, 119-127.
[71].Watada, J. (1992). Fuzzy time series analysis and forecasting of sales. Physica-Verlag, J. Kacprzyk, M. Fedrizzi, 211-227.
[72].Wong, J.-T., & Chung, Y.-S. (2007). Rough set approach for accident chains exploration. Accident Analysis and Prevention, 39, 629-637.
[73].Xu, B., Zhou, Y., & Lu, H. (2005). An improved accuracy measure for rough sets. Journal of Computer and System Sciences, 71 163-173.
[74].Yeh, C.-A. (2004). New Fuzzy time series approaches for forecasting in expenditure of information project. Unpublished thesis, National Yunlin University of Science and Technology, Touliu, Yunlin.
[75].Yu, H. K. (2005a). A refined fuzzy time series model for forecasting. Physica A, 346, 657-681.
[76].Yu, H. K. (2005b). Weighted fuzzy time series models for TAIEX forecasting. Physica A, 349, 609-624.
[77].Zadeh, L. A. (1965). Fuzzy sets. Information and Control, 8, 338-353.
[78].Zadeh, L. A. (1973a). The concept of a linguistic variable and its application to approximate reasoning. Berkeley: Memorandum ERL-M 411.
[79].Zadeh, L. A. (1988). Fuzzy logic. IE/EE Computing Magazine, 21(4), 83-93.
[80].Zadeh, L. A., Klir, G. J., & Yuan, B. (1996). Fuzzy sets, fuzzy logic, and fuzzy systems : selected papers. Singapore ; River Edge, N.J: World Scientific.
[81].Ziarko, W. (1994). Rough Sets, Fuzzy Sets and Knowledge Discovery. London: Springer.
[82].Ziarko, W. P., & Yao, Y. (2001). Rough sets and current trends in computing, second international conference, RSCTC 2000, Banff, Canada, October 16-19, 2000 : revised papers. from Connect to Lecture Notes in Computer Science http://gateway.library.qut.edu.au/login?url=http://www.springerlink.com/link.asp?id=105633
[83].Zimmermann, H. J. (2001). Fuzzy set theory--and its applications (4th ed.). Dordrecht ; Boston: Kluwer Academic Publishers.