穩健平均數–變異數模型與平均數-CVaR模型:越南股市投資組合的最優化應用

Robust Mean-Variance and Mean-CVaR in Portfolio Optimization: An application to Vietnamese stock market

士錦德(Cam-Duc Sy)

指導教授 ： 張健邦

朝陽科技大學/管理學院/企業管理系/碩士(2017年)

摘要

本研究主要是依據Markowitz (1952) 的現代投資組合理論建構越南股票市場從2015年2月到2017年2月期間的最優化投資組合。本研究完成平均數-CVaR模型及穩健平均數-變異數模型的實證分析並用以與資產報酬採用樣本平均數及變異數的傳統平均數-變異數模型比較。證實結果顯示， 穩健平均數-變異數投資組合的目標風險與VaR總是比傳統的投資組合低，夏普比率也不是例外。另外，實證結果也顯示在較高信賴水平下的平均數-CVaR投資組合回報較高的VaR和CVaR。

關鍵字

平均數-CVaR ； 穩健平均數－變異數 ； 投資組合最優化 ； 越南股市

並列摘要

This study aims to construct optimized portfolios of Vietnamese Stock Market based on Modern portfolio theory of Markowitz (1952) during the periods from Feb 2015 to Feb 2017. We provide empirical evidence by implement both Mean-CVaR model and robust Mean-Variance model with robust covariance estimators to compare with the traditional mean-variance model that using sample mean and covariance matrix of asset returns. , the target risks and VaR of robust mean-varian portfolios always are lower than the traditional model, the Sharpe ratio is not an exception. Besides, the empirical findings indicate the mean-CVaR portfolios under higher confidence level will return the higher VaR and CVaR.

並列關鍵字

Mean-CVaR ； Robust Mean-Variance ； Portfolio Optimization ； Vietnamese Stock Market

參考文獻

Ambachtsheer,K. (2005). Beyond portfolio theory: the next frontier. Financial Analysts Journal, Vol.6, No.1, pp.29-33.

Bartosz Sawik. (2010). Selected multiobjective methods for multiperiod portfolio optimization by mixed integer programming. In Kenneth D. Lawrence, Gary Kleinman (ed.), Applications in Multicriteria Decision Making, Data Envelopment Analysis, and Finance (Applications of Management Science, Vol 14), pp.3 – 34.

Bartosz T. Sawik. (2012). Conditional Value-at-Risk Vs Value-at-Risk to Multi-Objective Portfolio Optimization, Applications of Management Science, Vol.15, pp.277 – 305.

Bekkers, Niels and Doeswijk, Ronald Q. and Lam, Trevin W. (2009). Strategic Asset Allocation: Determining the Optimal Portfolio with Ten Asset Classes.

Bennett Mark Joseph. (2014). Data Mining with Markowitz Portfolio Optimization in Higher Dimensions.