IJPAM: Volume 25, No. 2 (2005)

SMOOTHING METHOD FOR A MINIMAX PROBLEM
IN PORTFOLIO SELECTION

Mingming Liu$^{1}$, Yan Gao$^{2}$
$^{1,2}$School of Management
University of Shanghai of Science and Technology
516 Jungong Road, Shanghai, 20093, P.R. CHINA
$^1$e-mail: sumoonlmm@sina.com.cn
$^{2}$e-mail: gaoyan1962@yahoo.com.cn


Abstract.In this paper, a minimax model for portfolio selection is presented. By maximum entropy technique, the nonsmooth minimax problem is approximated by a smooth problem. Based on KKT optimality conditions, this smooth optimization problem with equality constraints is transformed into a system of smooth equations. Then the classical Newton method is applied to solving it. The numerical test shows the valid of this method.

Received: October 10, 2005

AMS Subject Classification: 91B28, 90C47, 49K35

Key Words and Phrases: maximum entropy, minimax problem, optimization, portfolio selection

Source: International Journal of Pure and Applied Mathematics
ISSN: 1311-8080
Year: 2005
Volume: 25
Issue: 2