Multi-period portfolio optimization using coherent fuzzy numbers in a credibilistic environment

作     者：Gupta, Pankaj Mehlawat, Mukesh Kumar Khan, Ahmad Zaman 

作者机构：Univ Delhi Dept Operat Res Delhi 110007 India 

出 版 物：《EXPERT SYSTEMS WITH APPLICATIONS》 (专家系统及其应用)

年 卷 期：2021年第167卷

页      面：114135-114135页

核心收录：

学科分类：1201[管理学-管理科学与工程(可授管理学、工学学位)] 0808[工学-电气工程] 08[工学] 0812[工学-计算机科学与技术（可授工学、理学学位）] 

基　　金：MATRICS Scheme of DST-SERB, New Delhi, India Council of Scientific and Industrial Research - Human Resource Development Group (CSIR-HRDG) [09/045(1691)/2019-EMR-I] 

主　　题：Fuzzy multi-period portfolio optimization Multi-objective programming Fuzzy set theory Coherent fuzzy numbers Conditional Value-at-Risk Real-coded genetic algorithm 

摘      要：In this paper, we use an extension of fuzzy numbers, called coherent fuzzy numbers, to model asset returns and an investor s perception of the stock market (pessimistic, optimistic, or neutral) simultaneously. Two multi-period multi-objective portfolio optimization models are formulated using mean absolute semi-deviation and Conditional Value-at-Risk (CVaR) as risk measures, respectively. We aim to provide more flexibility to the investor in specifying the risk tolerance and devise optimum investment plans for different investment horizons. The proposed models also incorporate bound, cardinality, and skewness constraints for each investment period to capture various stock market scenarios. A real-coded genetic algorithm is employed to solve the resultant models. Two real-life case studies involving 20 assets of the National Stock Exchange (NSE), India, and another involving 50 assets listed in the S&P 500 and NASDAQ-100 indexes have been provided to illustrate the efficacy and advantages of the models. An in-sample and out-of-sample analysis have been done for both the models to analyze the performance in the real-world scenario. The conclusion drawn from the analysis strongly emphasizes on accurately assessing the current stock market prospects, i.e., adopting the right attitude (pessimistic, optimistic, or neutral), is of paramount importance and must be included in the portfolio optimization problem.