在非確定環境下最優投資組合策略; Optimal Portfolio Strategies Under the Environment with Uncertainties

Please use this identifier to cite or link to this item: http://ir.lib.ncu.edu.tw/handle/987654321/60924

Title:	在非確定環境下最優投資組合策略;Optimal Portfolio Strategies Under the Environment with Uncertainties
Authors:	秦承偉;Qin,Chengwei
Contributors:	統計研究所
Keywords:	不確定性;最佳投資策略;預測性回歸;HJB方程;貝氏分析;VAR模型;Uncertainties;Optimal portfolio strategy;Predictive regressions;HJB equations;Bayesian analysis;VAR model
Date:	2013-07-02
Issue Date:	2013-08-22 12:06:51 (UTC+8)
Publisher:	國立中央大學
Abstract:	投資者通常無法准確地把握投資期限，因為投資大多時候會因為某些意想不到的事情而終止，例如投資代理死亡。但有時他們會知道一些與投資期限相關的信息，並由此來決定投資策略。在這篇文章中，我們把重點放在分析和討論股票收益率預測性、模型參數不確定性和投資期限不確定性等因素對於投資策略的影響上。這個研究包含了兩個不同的投資方式。首先，我們通過一個離散的資產分配模型提出了買入持有策略，即在投資結束前不會改變資產分配權重；另一方面，對於資產分配權重隨?時間變動的動態模型，我們將用到連續時間模型。我們嘗試找出二者的最佳投資策略，從而使得在投資結束時所獲得的總收益達到最大。 Mostly, investors fail to accurately determine the length of the investment horizon. In many cases, the investment will stop because of some unexpected accidents such as the death of the agent. But sometimes they may have other informations related to the length of horizon which can give them an access to optimize their portfolio strategies. In this paper, we concentrate on investigating how will the portfolio strategies be e ected by the stochastic time-horizon and other two uncertainties including returns predictability and parameter uncertainty. Our research will involve two di erent investment philosophies. Firstly, we analyse the buy-and-hold strategy by a discrete-time model so that the portfolio choice can be static until the investment gets nished. On the other hand, we propose the continuous-process when considering the allocation weights are dynamic which means the strategies will be changed over time. Our primary purposes for both of them are trying to nd the optimal strategies which can maximize the wealth in terminal time.
Appears in Collections:	[Graduate Institute of Statistics] Electronic Thesis & Dissertation

Files in This Item:

File	Description	Size	Format
index.html		0Kb	HTML	655	View/Open

社群 sharing

Loading...