Volume 2, Number 2, May 2006, pp. 367-384
Rei Yamamoto and Hiroshi Konno
Key words:
mean-variance model, nonconvex transaction cost, linearization, 0-1 mixed integer programming problem
Mathematices Subject Classification: 90C11, 90C26, 90C90
ONLINE SUBSCRIPTION (Institutional Subscription Only)
Copyright© 2006 Yokohama Publishers
Back

Abstract:
This paper is concerned with an efficient algorithm for solving a mean-variance model under nonconvex transaction costs. We will show that this difficult problem can now be solved by using classical linearization technique of a separable nonlinear function. The resulting problem is a mixed zero-one integer programming problem with several thousand 0-1 integer variables, so that it is still beyond the reach of powerful integer programming softwares. However, by using a variety of heuristics, we can now solve a medium scale problem within a practical amount of time.
An efficient algorithm for solving a mean-variance model under nonconvex transaction costs

Special Issue of ICOTA6