在金融的領域中，選股早已經被認為是一個充滿挑戰和重要的任務。在近年來的發展中，機器學習和資料挖掘的技術提供一個契機，讓學者們能更進一步的研究和探討這類問題。因此在本篇研究中，我們以先前的單目標基因演算法選股為架構，改良並衍生到多目標基因模式，來更深入探討選股與投資組合的問題。我們首先針對股票的基本分析指標進行特徵選取與權重最佳化，透過特徵選取與最佳化後的權重來針對股票進行評分並排名，選擇排名較佳的股票組成投資組合，不同的染色體將會形成不同的投資組合，比較不同投資組合的報酬與風險，並透過多目標基因演算法來選出較佳的投資組合。其次，我們進一步改良多目標基因演算法，加入財務領域的知識來協助投資組合選擇的判斷。除此之外，我們在多目標基因演算法模型的基礎上加入資產配置的概念，將資產分配到不同類股上。依據我們所獲得的良好實驗結果，我們期許本研究的方法能推動軟式計算技術在真實金融股票投資上的應用。

Stock selection has long been recognized as a challenging and important task in finance. Recent advances in machine learning and data mining are leading to significant opportunities to solve these problems more effectively. In this study, we enrich our work for stock selection using single-objective genetic algorithms (SOGA) and extend it to multi-objective GA (MOGA) models. We first employ the SOGA for optimization of model parameters and feature selection for input variables to the model, and then devise a stock scoring mechanism to rank and select stocks for forming a portfolio. With each chromosome representing a feasible portfolio, the adopted MOGA models thus decide good portfolios by considering their return and risk. We also improve upon the MOGA models using financial knowledge to help selection of beneficial portfolios. Furthermore, we present an investigation for asset allocation in various industrial sectors using our proposed models. Based on the promising results, we expect this MOGA methodology to advance the current state of research in soft computing for real-world stock selection applications.

論文審定書 i
致謝 ii
摘要 iii
Abstract v
Contents vii
List of Figures ix
List of Tables x
CHAPTER 1. Introduction 1
1.1. Motivation 1
1.2. Contributions 3
1.3. Organization of Thesis 4
CHAPTER 2. Related Works 5
2.1. Genetic Algorithm 5
2.1.1. Model Optimization for Stock Selection by the Single-objective Genetic Algorithms (SOGA) 8
2.2. Multi-objective Algorithm 12
CHAPTER 3. Problem Statement and Definitions 15
CHAPTER 4. Proposed Algorithms 24
4.1. Model Optimization by the Multi-objective Genetic Algorithm (MOGAold)... 24
4.2. Model Optimization by the Modified Multi-objective Genetic Algorithm (MOGAnew) 29
4.3. Model Optimization and Assets Allocation by Modified Multi-objective Genetic Algorithm (MOGAnew-a) 31
CHAPTER 5. Datasets 33
5.1. Stocks of the 200 Largest Market Capitalizations 33
5.2. Stocks of Various Sectors 33
CHAPTER 6. Experimental Results and Statistical Validation 35
6.1. Comparison of the Experimental Results 36
6.1.1. Comparison of the Benchmark, SOGA, MOGAold, and MOGAnew... 36
6.1.2. Comparison of the Benchmark, MOGAold, and MOGAnew for Stock Selection in Five Sectors 42
6.1.3. Comparison of the Benchmark, MOGAnew-a, and Markowitz’s Model in Stocks in Five Sectors 58
CHAPTER 7. Conclusions and Future Works 60
References 62

[1] Y. L. Becker, P. Fei, and A. Lester, “Stock Selection: An Innovative Application of Genetic Programming Methodology,” Genetic Programming Theory and Practice IV, Genetic and Evolutionary Computation, pp. 315-334, 2007.
[2] J. D. Bermudez, J. V. Segura, and E. Vercher, “A Fuzzy Ranking Strategy for Portfolio Selection Applied to the Spanish Stock Market,” The 2007 IEEE International Conference on Fuzzy Systems, pp.1-4, 2007.
[3] N. Chapados and Y. Bengio, “Cost Functions and Model Combination for VaR-Based Asset Allocation Using Neural Networks,” The 2001 IEEE Transactions on Neural Networks, Vol. 12, No. 4, pp. 890-906, 2001.
[4] C. H. Chiang, Using Genetic Algorithms to Find Fundamental Selection and Technical Timing Rules, Master Thesis, National Central University, 2002.
[5] K. Deb, A. Pratap, S. Agarwal, and T. Meyarivan, “A Fast and Elitist Multiobjective Genetic Algorithm: NSGA-II,” The 2002 IEEE Transactions on Evolutionary Computation, Vol. 6, No.2, pp. 182-197, 2002.
[6] G. Hassan and C. Clack, “Robustness of Multiple Objective GP Stock-Picking in Unstable Financial Markets: Real-Word Applications Track,” The 11th Annual Conference on Genetic and Evolutionary Computation, pp. 1513-1520, 2009.
[7] Hoklie and L.R. Zuhal, “Resolving Multi Objective Stock Portfolio Optimization Problem Using Genetic Algorithm,” The Second International Conference on Computer and Automation Engineering, Vol. 2, pp. 40-44, 2010.
[8] J. H. Holland, Adaptation in Natural and Artificial Systems, University of Michigan Press, 1975.
[9] C. F. Huang, “A Hybrid Stock Selection Model Using Genetic Algorithms and Support Vector Regression,” Applied Soft Computing, Vol. 12, No. 2, pp. 807-818, 2012.
[10] C. F. Huang, C. H. Chang, B. R. Chang, and D. W. Cheng, “A Study of a Hybrid Evolutionary Fuzzy Model for Stock Selection,” The 2011 IEEE International Conference on Fuzzy Systems, pp. 210-217, 2011.
[11] C. F. Huang, B. R. Chang, D. W. Cheng and C. H. Chang, “Feature Selection and Parameter Optimization of a Fuzzy-based Stock Selection Model using Genetic Algorithms,” International Journal of Fuzzy Systems, Vol. 14, No. 1, pp. 65-75, 2012.
[12] K. J. Kim and I. Han, “Genetic Algorithms Approach to Feature Discretization in Artificial Neural Networks for the Prediction of Stock Price Index,” Expert Systems with Applications, Vol. 19, No. 2, pp. 125-132, 2000.
[13] K. K. Lai, L. Yu, S. Wang, and C. Zhou, “A Double-Stage Genetic Optimization Algorithm for Portfolio Selection,” The 13th International Conference on Neural Information Processing, Lecture Notes in Computer Science, Vol. 4234, pp. 928-937, 2006.
[14] Y. F. Li and W. Guo, “The Stock Portfolios Simulated Annealing Genetic Algorithm Based on RAROC,” The 21st Chinese Control and Decision Conference, pp. 3574-3578, 2009.
[15] J. Li and J. Xu, “Multi-Objective Portfolio Selection Model with Fuzzy Random Returns and a Compromise Approach-Based Genetic Algorithm,” Information Sciences, Vol. 220, pp. 507-521, 2013.
[16] D. Lohpetch and D. Corne, “Multiobjective Algorithms for Financial Trading: Multiobjective Out-Trades Single-Objective,” The 2011 IEEE Congress on Evolutionary Computation, pp. 192-199, 2011.
[17] Y. Orito and G. Yamazaki, “Index Fund Portfolio Selection by Using GA,” The 2001 IEEE International Conference on Computational Intelligence, pp.118-122, 2001.
[18] R. Pereira, Genetic Algorithm Optimisation for Finance and Investments, Discussion Paper, La Trobe University, 2000.
[19] T. S. Quah and B. Srinivasan, “Improving Returns on Stock Investment through Neural Network Selection,” Expert Systems with Applications, Vol. 17, No. 4, pp. 295-301, 1999.
[20] A. Sülflow, N. Drechsler, and R. Drechsler, “Robust Multi-Objective Optimization in High Dimensional Spaces,” The 4th International Conference on Evolutionary Multi-Criterion Optimization, Lecture Notes in Computer Science, Vol. 4403, pp. 715-726, 2007.