多情境資產配置系統雛形之建置

Multi-scenario asset allocation system

李青憲(Ching-Hsien Lee)

指導教授 ： 盧以詮

元智大學/資訊學院/資訊管理學系/碩士(2002年)

摘要

傳統上進行長期資產配置時，常以平均數變異數架構來進行分析，其模式投入要素如資產的預期報酬率、變異數等是以由歷史平均值所形成。然而，由於投資工具之歷史報酬率反映未來能力不佳，且不同時期所選出的歷史報酬率彼此差異極大，造成所產生的效率前緣呈現出不穩定。有鑑於此，本論文將以多種情境模式來進行資產配置，除了可以考慮到未來各種可能產生的經濟情況及其機率外，並預估各種情境下不同資產工具之報酬率，使得最後所求得之效率前緣可以與未來實際狀況更加接近。因此本論文以多情境資產配置流程的為主軸，建構分散式多情境資產配置系統平台，藉此讓使用者可以根據系統平台所提供各項資訊，快速建構更有效率的資產配置。

關鍵字

多情境資產配置 ； 分散式系統

並列摘要

In order to proceed the result of the long term asset allocation, we often use Mean-Variance optimization (MVO) model. According to the related thesis about MVO model, forecasts of the expected returns and covariance of the assets are on the basis of historical average at some period in the past. However, historical average is not sufficient to present the conditions and historical averages at some different period time, so the results of the efficient frontier are huge different and unstable.Therefore, this thesis using multi-scenarios model for asset allocation not only takes into consideration about a large number of potential scenarios and every scenario’s probability, but also estimates the different asset’s return in every scenario. By this way, it will make the efficient frontier more correct. Finally, we propose a distributed multi-scenario asset allocation system providing information for investors to construct the asset allocation efficiently.

並列關鍵字

Multi-Scenario Asset Allocation ； Distributed System

參考文獻

[5]Brinson, Gary P., Brian D. Singer and Gilbert L. Beebower, “Determinants of Portfolio Performance Ⅱ:An Update,”Financial Analysts Journal,May/June 1991.

[7]Markowitz, Harry M.“Portfolio Selection,” Jouranl of Finance, March ,1952

[8]Y. K. Koskosidis, “A Scenario-Based Approach to Active Asset Allocation,” Journal of Portfolio Management, 1997.

[9]Michael Edesess, Heoge A. Hambrecht, “Scenario Forcasting; Necessity, not choice,”Financial Analyst Journal, 1990.

[10]Martin L. Leibowitz, Roy D. Herriksson, “Portfolio optimization with shortfall constraints : a confidence-limit approach to managing downside risk,” Financial Analyst Journal, Mar-Apr 1989.