A Study on Fuzzy Entropy Model in Portfolio Selection

碩士 === 國立暨南國際大學 === 資訊管理學系 === 102 === Abstract On the development of portfolio selection, Entropy is not only dealt with the issue of non-diversification but used as the measurement of risk to replace the variance in Mean-Variance portfolio selection model. Besides, to describe the uncertainty of f...

Full description

Bibliographic Details
Main Authors: Yi-Wei Chen, 陳奕維
Other Authors: Jing-Rung Yu
Format: Others
Language:zh-TW
Published: 2014
Online Access:http://ndltd.ncl.edu.tw/handle/72627130162801045046
Description
Summary:碩士 === 國立暨南國際大學 === 資訊管理學系 === 102 === Abstract On the development of portfolio selection, Entropy is not only dealt with the issue of non-diversification but used as the measurement of risk to replace the variance in Mean-Variance portfolio selection model. Besides, to describe the uncertainty of future return (fuzzy return) and the range of the return that the investors can accept, Fuzzy theory is used to construct portfolio selections under the fuzzy environment. Huang (2008) proposed Fuzzy-Mean-Entropy model (FME) and considered the elements of returns, risk, fuzzy variables and the entropy risk measurement. Due to FME model still have the issue with MV model of non-diversification, we use Entropy to handle two kinds of issue of diversification and the uncertainty of future return, to construct the portfolio for maximal returns and minimal risk. Under the fuzzy environment we consider the Entropy model proposed by Yager (1995) and construct the two objectives Fuzzy Mean_Yager-entropy model (FM_Y) to deal with fuzzy return and risk diversified. On the basis of FME model, we integrate Yager’s Entropy model to construct the three objectives Fuzzy Mean-Entropy_Yager-entropy (FME_Y) afterwards. And compare with the 1/N (Buy and Hold) model, MV model and FME model, we use the performance measures of portfolio such as Market value, Sharpe ratio, Omega ratio etc. to analyze each model. The research shows that (1) FME model with fuzzy returns and minimal uncertain returns can measure the uncertain returns and risk more accurate than MV model. (2) The FME_Y model considers Fuzzy entropy and risk diversified at the same time can maintain less loss and higher benefit during the down turns of economy cycles.