Network techniques for solving asset diversification problems in finance

In this paper, we study a problem of asset diversification or dynamic portfolio selection arising in the field of finance. This problem is first formulated as a nonlinear mathematical program and is solved via a special-purpose computer code that exploits the underlying network structure. A second m...

Full description

Saved in:
Bibliographic Details
Published inComputers & operations research Vol. 9; no. 3; pp. 173 - 195
Main Authors Golden, Bruce L., Keating, Karen D.
Format Journal Article
LanguageEnglish
Published New York Elsevier Ltd 1982
Pergamon Press Inc
Subjects
Assets
Computer based modeling
Diversification
Mathematical programming
Network management systems
Networks
Operations research
Portfolio management
Problem solving
Techniques
Online AccessGet full text

Cover

More Information
Summary:In this paper, we study a problem of asset diversification or dynamic portfolio selection arising in the field of finance. This problem is first formulated as a nonlinear mathematical program and is solved via a special-purpose computer code that exploits the underlying network structure. A second model for asset diversification is presented which is linear in the decision variables and is therefore much easier to solve computationally than the first formulation. On the other hand, the linear model produces solutions that are consistently inferior to those generated by the nonlinear program in terms of expected profit. This trade-off of accuracy versus efficiency is discussed and assessed for several sample problems. Extensive computational results are presented.
ISSN:0305-0548
1873-765X
0305-0548
DOI:10.1016/0305-0548(82)90017-X