臺灣博碩士論文加值系統

儘管傳統Markowitz均數變異數分析已成為現代財務理論的發展過程中主要的理論之一，但卻未能廣泛的應用為股權管理的實際工具。Markowitz最適投資組合迷思的主要原因在於對輸入參數估計時所產生的嚴重風險，而導致不具財務內涵或是錯誤的組合及資產配置的建議。因此，本研究提出了一個拔靴類神經網路的投資組合法來改善組合構建之績效。詳言之，將殘差拔靴法應用於多層前向式神經網路預測模型，建立再抽樣以估算出組合之預期收益和共變異矩陣，然後再將其導入傳統Markowitz最適化過程。本研究比較本拔靴類神經網路的投資組合、傳統的Markowitz的均數變異數分析、James-Stein及minimum-variance估計之投資績效。實證結果發現，拔靴類神經建構投資組合優於其他三種基準模式，在所有設定的績效指標下均擁有較高報酬。本研究提供之證據均說明了此新組合構建程序確實顯著加強了Markowitz均數變異數分析的投資價值。

ABSTRACT
Despite having become firmly established as one of the major cornerstone principles of modern finance, traditional Markowitz mean-variance analysis has, nevertheless, failed to gain widespread acceptance as a practical tool for equity management. The Markowitz optimization enigma essentially centers on the severe estimation risk associated with the input parameters, as well as the resultant financially irrelevant or even false optimal portfolios and asset allocation proposals. We therefore propose a portfolio construction method in the present study which incorporates the adoption of bootstrapping neural network architecture. In specific terms, a residual bootstrapping sample, which is derived from multilayer feedforward neural networks, is incorporated into the estimation of the expected returns and the covariance matrix, which are then, in turn, integrated into the traditional Markowitz optimization procedure. The efficacy of our proposed approach is illustrated by comparing it with traditional Markowitz mean-variance analysis, as well as the James-Stein and minimum-variance estimators, with the empirical results indicating that this novel approach significantly outperforms the benchmark models, in terms of various risk-adjusted performance measures. The evidence provided by this study suggests that this new approach has significant promise with regard to the enhancement of the investment value of Markowitz mean-variance analysis.

CONTENT LIST
Essay I. Portfolio Construction Using Bootstrapping Neural Networks: Evidence from Taiwan Stock market
摘要 i
ABSTRACT ii
誌謝 iv
CONTENT LIST v
TABLE LIST viii
FIGURE LIST ix
緒論 1
THESIS PREFACE 2
1. INTRODUCTION 5
2. DATA AND METHODOLOGY 11
2.1 Multilayer Feedforward Neural Networks 11
2.2 Mean-Variance Optimization 12
2.3 James-Stein and Minimum-variance Estimations 13
2.4 Portfolio Construction Using Bootstrapping Neural Networks 14
2.5 Empirical Data and Methodology 16
3. EMPIRICAL RESULTS AND ANALYSIS 19
3.1 The Sharpe Ratio 21
3.2 The Risk Adjusted Returns 24
4. CONCLUSIONS 29
References 30
Essay II. Portfolio Construction Using Bootstrapping Neural Networks: Evidence from Global Stock markets
摘要 34
ABSTRACT 35
1. INTRODUCTION 37
2. DATA AND METHODOLOGY 43
2.1 The Elman Network 43
2.2 Mean-Variance Optimization 44
2.3 Portfolio Construction Using Bootstrapping Neural Networks 45
2.4 Empirical Data and Methodology 47
3. EMPIRICAL RESULTS AND ANALYSIS 49
3.1 The Sharpe Ratio 52
3.2 The Risk Adjusted Returns 55
4. CONCLUSIONS 59
APPENDIX 1 65
APPENDIX 2 67

References
[1] Board, J. and C. Sutcliffe, 1994, “Estimation Methods in Portfolio Selection and the Effectiveness of Short Sales Restrictions: UK Evidence”, Management Science, vol. 40(4), pp.516-534.
[2] Efron, B. and R. Tibshirani, 1993, An Introduction to the Bootstrap, New York, NY: Chapman and Hall.
[3] Elman, J.L., 1990, “Finding Structure in Time”, Cognitive Science, vol. 14, pp.179-211.
[4] Fernandez, A. and S. Gomez, 2007, “Portfolio Selection using Neural Networks”, Computers and Operations Research, vol. 34, pp.1177-1191.
[5] Franke, J. and K. Matthias, 1999, Optimal Portfolio Management using Neural Networks: A Case Study, Kaiserslautern, Germany: University of Kaiserslautern, Department of Mathematics Technical Report.
[6] Franke, J. and M. Neumann, 2000, “Bootstrapping Neural Networks”, Neural Computation, vol. 12(8), pp.1929-1949.
[7] Freedman, D.A., 1982, “Bootstrapping Regression Models”, Annual Statistics, vol. 9, pp.1218-1228.
[8] Frost, P.A. and J.E. Savarino, 1986, “An Empirical Bayes Approach to Efficient Portfolio Selection”, Journal of Financial and Quantitative Analysis, vol. 21(3), pp.293-305.
[9] Graham, J. R. and Harvey C. R., 1997, “Grading the Performance of Market-Timing Newsletters”, Financial Analysts Journal, vol. 53(6), pp.54-66.
[10] Haykin, S., 1999, Neural Networks: A Comprehensive Foundation, Upper Saddle River, NJ: Prentice-Hall Inc.
[11] Hornik, K., Stinchcombe, M., and White, H., 1989, “Multilayer Feedforward Networks Are Universal Approximators”, Neural Networks, vol. 2, pp.359-366.
[12] Jobson, J.D. and B. Korkie, 1981, “Putting Markowitz Theory to Work”, Journal of Portfolio Management, vol. 7(4), pp.70-74.
[13] Jorion, P., 1986, “Bayes-Stein Estimation for Portfolio Analysis”, Journal of Financial and Quantitative Analysis, vol. 21(3), pp.279-292.
[14] Lin, C.M., J.J. Huang, M. Gen and G.H. Tzeng, 2006, “Recurrent Neural Networks for Dynamic Portfolio Selection”, Applied Mathematics and Computers, vol. 175, pp.1139-1146.
[15] Markowitz, H.M., 1952, “Portfolio Selection”, Journal of Finance, vol. 7(1), pp.77-91.
[16] Markowitz, H.M., 1959, Portfolio Selection: Efficient Diversification of Investments, New York, NY: John Wiley & Sons.
[17] Michaud, R.O., 1989, “The Markowitz Optimization Enigma: Is Optimized Optimal”, Financial Analysts Journal, vol. 45, pp.31-42.
[18] Michaud, R.O., 1998, Efficient Asset Management: A Practical Guide to Stock Portfolio Optimization and Asset Allocation, Boston, MA: Harvard Business School Press.
[19] Modigliani F. and Modigliani L., 1997, “Risk-Adjusted Performance: How to Measure It and Why”, Journal of Portfolio Management, vol. 23(2), pp.45-54.
[20] Refenes, A.N., A.D. Zapranis, J.T. Connor and D.W. Bunn, 1994, “Neural Networks in Investment Management”, Journal of Communication and Finance, vol. 8, pp.95-101.
[21] Sharpe, W., 1963, “A Simplified Model for Portfolio Analysis”, Management Science, vol. 9, pp.277-293.
[22] Stein, C., 1955, “Inadmissibility of the Usual Estimator for the Mean of a Multivariate Normal Distribution”, Proceedings of the Third Berkeley Symposium on Mathematical Statistics and Probability, pp.197-206.
[23] Treynor, J. and F. Black, 1973, “How to Use Security Analysis to Improve Portfolio Selection”, Journal of Business, vol. 46, pp.66-86.
[24] White, H., 1989, “Some Asymptotic Results for Learning in Single Hidden-layer Feed-forward Network Models”, Journal of American Statistical Association, vol. 84, pp.1008-1013.
[25] Yu, L., S. Wang and K.K. Lai, 2008, “Neural Network-based Mean-variance Skewness Model for Portfolio Selection”, Computers and Operations Research, vol. 35, pp.34-46.