臺灣博碩士論文加值系統

平均數／變異數分析向來是進行長期資產配置時最常用的數量方法。在傳統分析架構裡，投入要素如資產的預期報酬率、變異數等乃是由歷史平均值所形成。然而，當經濟環境或金融市場有重大改變時，歷史平均值將無法正確地反映目前情況。有鑑於此，本論文的研究重點乃在於探討以歷史平均報酬率形成預期的缺失，並且嘗試提出另一套能夠考慮未來多種可能情境發生的資產配置流程。
本篇論文首先以國內投資工具為實證研究的對象，並且以傳統取平均值之方法形成預期報酬率等投入要素來進行資產配置，結果發現投資工具之歷史報酬率反映未來的能力不佳，且不同時期所選出的歷史報酬率彼此差異極大；而在此傳統方法之下，效率前緣更呈現出不穩定的問題。
不同於傳統分析架構，多種情境模式的資產配置考慮到未來各種可能產生的經濟情況，並預估各種情況發生的機率，使得最後所求得之效率前緣可以與未來實際狀況更加貼近。進行多種情境模式之資產配置的過程當中，將會有遭遇幾項難題：首先是未來情境如何定義；其次是如何估計各情境發生的機率，最後是如何建立各情境之下投資工具之預期報酬率。這些問題亦將於內文中有詳細之闡述。

Mean-Variance optimization is the most common quantitative methodology employed for asset allocation. Under traditional mean-variance optimization techniques, forecasts of the expected return, expected volatility, and expected covariance of the assets are driven by consideration of the average market behavior at some period in the past, which period is chosen more or less arbitrarily. One drawback of this approach is that as world economic conditions and financial markets change, historical averages may not be sufficiently representative of current conditions. Therefore, this thesis sets out as its primary goal to discuss the effects of estimation bias on asset allocation using traditional methods; this study also presents an improved mean-variance optimization technique which takes into consideration a large number of potential return outcomes.
We test the effects of estimation bias on asset allocation utilizing traditional methods with domestic assets. We find, however, that the historical patterns of assets are generally unable to anticipate correctly assets'' future moves, especially stock returns. Besides, the average returns chosen from different time periods differ considerably, resulting in unstable efficient frontiers.
Scenario-based asset allocation provides the means to overcome these shortcomings. Instead of relying on a single-point forecast, which is in essence a single scenario, through scenario-based asset allocation investors can provide a set of plausible scenarios of future expected returns and diversify their portfolio by taking into account potential return outcomes. In the course of this research we encountered some difficulties, such as the exact definition of future scenarios, or the estimation of probability that certain scenario will happen; these difficulties, however, are discussed in detail and practicable solutions are put forward.

第一章 緒論……………………………………………...5
第一節 研究動機…………………………………..…………….5
第二節 研究目的7
第三節 研究架構8
第二章 決定最適資產配置文獻探討9
第一節 平均數／變異數架構相關研究9
第二節 決定最適資產配置之方法13
第三章 傳統求算效率前緣方法之缺失15
第一節 以歷史報酬率形成預期求算效率前緣的問題15
第二節 傳統方法缺失之實證研究16
第三節 改善方法23
第四章 多種情境模式之分析27
第五章 個案研討43
第一節 國內開放式共同基金43
第二節 全球股票型基金59
第六章 結論78
參考文獻81

中文部份
1. 閔志清，「台灣基金資產配置之研究」，台大財研碩士論文，民國八十七年六月
2. 陳玫纓，「台灣退休基金資產配置與投資組合保險策略之研究」，台大財研碩士論文，民國八十六年六月
3. 郭自恭，「日本之財政赤字及其重建」，國際際金融參考資料第三十九輯，民國八十五年六月
4. 陳松男，「國際金融市場泛論與分析」，台北：新陸書局，民國八十五年一月
5. 邱靖博、蔡豐清，「金融市場」，第二版，台北：三民書局，民國八十二年九月
英文部份
1. Arnott, Robert D. , 1985 , The Pension Sponsor’s View of Asset Allocation. Financial Analysts Journal, Sep.-Oct..
2. Brinson, Gary P., Brian D. Singer and Gilbert L. Beebower, 1991, Determinants of Portfolio Performance II: An Update. Financial Analysts Journal, May-June.
3. Tsuno, Masanori, 1995, Security versus Returns in Japanese Pension Fund Management. OECD Paris Conference Pension Financing and Capital Formation, Feb.
4. Black, F., and Litterman. 1991a Asset Allocation：Combing Investors’ View with Market Equilibrium. Journal of Fixed Income, September.
5. Markowitz, Harry M., 1952, Portfolio Selection. Journal of Finance, March
6. William S. Gray，1993, Historical Returns，”Inflation and Future Return Expectations. Financial Analyst Journal ，July-August
7. Vijay K. Chopra and William T. Ziemba，1993, The effect of errors in Means，Variances and Covariances on Optimal portfolio Choice. The Journal of Portfolio Management，winter
8. Yoav Benari，1990, Optimal asset mix and its link to changing fundamental factors. The Journal of Portfolio Management, Winter
9. Richard O. Michaud，1989, The Markowitz Optimization Engima：Is Optimized Optimal?. Financial Analyst Journal , Jan-Feb
10. Michael Edesess and Heoge A. Hambrecht，1990, Scenario Forecasting；Necessity，not choice. Financial Analyst Journal, Spring
11. Grant Mcqueen and Steven Thorley，1991, Are Stock Returns Predictable? A Test Using Markov Chains. The Journal of Financial ，Vol. XLVI，No 1，March
12. John M. Mulvey and Hercules Vladimirou , 1989, Stochastic network optimization models for investment planning. Annals of Operations Research，20
13. Martin L. Leibowitz and Roy D. Henriksson , 1989 , Portfolio optimization with shortfall constraints : a confidence-limit approach to managing downside risk. Financial Analyst Journal，Mar-Apr
14. Raj Aggarwal and Andrea L. Demaskey , 1997, Cross-hedging currency risks in Asian emerging markets using derivatives in major currencies. The Journal of Portfolio Management，Spring
15. Robert R. Grauer and Nils H. Hakansson , 1982, Higher returns , lowr risk: historical returns on long-run , actively managed portfolios of stocks , bonds and bills, 1936-1978. Financial Analyst Journal，March/April
16. Roger G. Clarke and Harindra de Silva , 1998, State-dependent asset allocation. The Journal of Portfolio Management，Winter
17. Brianton Geoffrey , 1997, Risk Management and Financial Derivatives p431-469,
18. Yiannis A. Koskosidis and Antonio M. Duarte , 1997, A scenario-based approach to active asset allocation, The Journal of Portfolio Management，Winter