English  |  正體中文  |  简体中文  |  Items with full text/Total items : 2737/2828
Visitors : 280127      Online Users : 14
RC Version 4.0 © Powered By DSPACE, MIT. Enhanced by NTU Library IR team.
Scope Adv. Search
LoginUploadHelpAboutAdminister

Please use this identifier to cite or link to this item: http://ir.lib.stu.edu.tw:80/ir/handle/310903100/1067

Title: 不同風險預測模式之投資組合績效比較 - 以國際資產配置為例
The Performance of Portfolio under Different Risk Measures - A study of International Assets Allocation
Authors: 洪慶昇
Ching-Sheng Hung
Contributors: 許江河
Philip Hsu
金融保險研究所
Keywords: 風險預測模型;指數平滑法;GARCH模型;最適投資組合;資產配置;拔靴複製法
Risk Predict Model;Exponential Smoothing;GARCH model;Optimal Portfolio;Asset Allocation;bootstrap method
Date: 2004
Issue Date: 2011-05-24 13:25:06 (UTC+8)
Publisher: 高雄市:[樹德科技大學金融保險研究所]
Abstract: 變異數是決定投資組合績效之關鍵因素,因此若能掌握標的股價指數之波動變化,即能透過變異數之預測進行資產配置,進而協助投資者決策之擬定。本文以摩根史坦利公司(MSCI)所編製之太平洋地區等五個國家股價指數為研究對象,分別採用傳統變異數法、固定視窗法、指數平滑法、GARCH模型等四種風險預測方法,估計個別股價指數之報酬率變異數,再將不同模型所求得變異數預測值,藉以考量最小變異數下之投資組合,進而研擬出最適合投資者所採用之資產配置策略。
本文以兩方面來探討透過何種風險預測模式之建構國際投資組合時,將得到較佳的投資組合績效:(一)在給定目標報酬率下,考慮匯率風險與區分情境之因素,透過效率前緣的求算,選擇風險最小變異數之投資組合,分析在限制條件下,對投資組合所造成的差異。(二)透過給定固定投資組合風險,利用預測風險模型所配置之權重,再乘上每個國家當期實際報酬,透過求得最高期望報酬率的觀點,驗證資產配置是否為最佳的投資決策,而獲得以下主要研究實證結果:
(一)由實證結果顯示,最適之投資組合配置的決定對不同預測方法的選取是敏感的,亦即當報酬率波動改變時,會因不同假設及方法論而有差異,進而影響資產配置,導致投資組合的績效有顯著不同。
(二)由實證分析的結果,在考慮匯率風險與區分情境之因素下,發現當報酬率波動改變時,依據樣本的特性來選擇合適波動性預測模型,透過指數平滑法來進行變異數的預測,其建立之資產配置,所形成的投資組合決策較能得到良好的績效。
(三)由實證結果發現,不論是透過何種風險預測模型,所預測出來的風險可得知,投資於國際證券市場時,價格風險才是最大的風險來源。倘若再考量避險成本,對國外資產進行匯率避險可能並非合適策略。
(四)本文建構與時間相依之GARCH(1,1)風險預測模型,其所提供之配置決策於給定限制條件下,並無法優於指數平滑法所配置之績效。
(五)本文於實證資料可得知,當給定投資組合報酬時,無論給定期間在任何情境,依各模型是否進行規避匯率風險而予以調整投資組合,並不會改變其優劣關係。
(六)本文於實證資料可得知,當投資組合風險固定時,藉由指數平滑法所配置之權重,乘於每個標的之當期實際報酬率。在空頭時期,不論探討避險與否,均可得較佳之投資組合績效。
上述之實證結果,可找到具有良好預測風險能力之模型,進而提供投資決策者掌握金融資產報酬波動性的可行管道。
Variance is the critical factor to decide the investment combination efficiency. If we hold the variance of stock index, we can do asset allocation through the prediction variance and help to make the investment policy further.
The object of this research is the stock index of five countries which are organized to Pacific Zone by Morgan Stanley Capital International Company ( MSCI ). To estimate individual reward variance by using four different risk predict models-Traditional Variance, Fixed Windows, Exponential Smoothing, Generalized Autoregressive Conditional Heteroskedasticity (GARCH). Then, to use the variance forecast values and do asset allocation, to get minimal variance portfolio.
This research that explores what kind of risk predict model can make the better effect than others from two aspects on setting International Portfolio:
(1) Under the specific target return, considering the exchanges, and analyzing the different situations, through the counting of efficient frontier to choose the investment combination of minimal risk variance. By analyzing the restricted conditions and the variance of investment combination, to make the asset allocation policy that adopted by International Fund Manager.
(2) Under the same investment combination risk, to use the weight that is distributed from risk predict model to multiply the present real return of every country and try to find the highest real return in the known risk situations to prove if the asset allocation is the best investment policy or not.
I expect that we can find good predict risk model and it can provide the possible way for investors to hold the fluctuation of finance asset rewards. The conclusion of research:
(1) The conclusion of experiment shows that the decision of Optimal Portfolio Asset Allocation to the choice of different risk models is closely. It also means that when the return variance is changed, there must be different because of suppositions and methodologies. It can affect the asset allocation further and the different performance of portfolio.
(2) The conclusion of analysis is that it can get better performance when we find the return variance is changed, to choose the suitable variance predict model according to the quality of sample and use Exponential Smoothing to proceed the predict of variance under considering exchange rate risks and distribution scenarios.
(3) We can find something from the conclusion of experiment, no matter what kind of risk predict models that we adopted, we can learn when investing to International Stock Market, the price risk is the largest risk. If considering hedging costs, it’s not a suitable way for hedge exchanged risk of foreign Asset.
(4) This chapter shows the allocation of GARCH(1,1) is not more superior than the performance of Exponential Smoothing.
(5) We can learn the information from data of evidence, when the portfolio risk is fixed, whether we proceed the hedge exchanged risk and adjust the portfolio or not, it can’t change the relationship between good and bad for any situations.
(6) We can learn the information from data of evidence, when the portfolio risk is fixed, every real return in time multiplied by the weight of Exponential Smoothing. In bear market, whether to discuss the hedge or not, it can get better the performance of portfolio.
Appears in Collections:[金融與風險管理系(所)] 博碩士論文

Files in This Item:

File Description SizeFormat
不同風險預測模式之投資組合績效比較 - 以國際資產配置為例__臺灣博碩士論文知識加值系統.htm國圖140KbHTML399View/Open


All items in STUAIR are protected by copyright, with all rights reserved.

 


無標題文件

著作權政策宣告:

1.

本網站之數位內容為樹德科技大學所收錄之機構典藏,無償提供學術研究與公眾教育等公益性使用,惟仍請適度,合理使用本網站之內容,以尊重著作權人之權益。商業上之利用,則請先取得著作權人之授權。
 
2. 本網站之製作,已盡力防止侵害著作權人之權益,如仍發現本網站之數位內容有侵害著作權人權益情事者,請權利人通知本校護人員(clairhsu@stu.edu.tw),維護人員將立即採取移除該數位著作等補救措施。
 
DSpace Software Copyright © 2002-2004  MIT &  Hewlett-Packard  /   Enhanced by   NTU Library IR team Copyright ©   - Feedback