Minimax portfolio optimization: empirical numerical study

• Published in: The Journal of the Operational Research Society 2004-01, Vol.55 (1), p.65-72
• Main Authors: Cai, X, Teo, K L, Yang, X Q, Zhou, X Y
• Format: Article
• Language: English
• Subjects: Applied sciences; Blue chip stocks; Business and Management; Exact sciences and technology; Expected returns; Experiments; Financial investments; Financial portfolios; Investment portfolios; Investment return rates; Investment risk; Investments; Management; numerical study; Operational research and scientific management; Operational research. Management science; Operations research; Operations Research/Decision Theory; Optimization; Optimization algorithms; Parametric models; Performance evaluation; Portfolio investments; Portfolio management; portfolio selection; Portfolio theory; Prices; Random variables; Rates of return; Risk aversion; risk aversion measures; standard deviation; Stock exchanges; Stocks; Studies; Theoretical Paper; Theoretical Papers; variance
• ISSN: 0160-5682; 1476-9360
• DOI: 10.1057/palgrave.jors.2601648

In this paper, we carry out the empirical numerical study of the l ∞ portfolio selection model where the objective is to minimize the maximum individual risk. We compare the numerical performance of this model with that of the Markowitz's quadratic programming model by using real data from the Stock Exchange of Hong Kong. Our computational results show that the l ∞ model has a similar performance to the Markowitz's model and that the l ∞ model is not sensitive to the data. For the situation with only two assets, we establish that the expected return of the minimum variance model is less than that of the minimum l ∞ model when both variance and the return rate of one asset is less than the corresponding values of another asset.