On convex multiobjective programs with application to portfolio optimization

Abstract

Focusing on (strictly) convex multiobjective programs (MOPs), we review some well‐established scalarizations in multiobjective programming from the perspective of parametric optimization and propose a modified hybrid scalarization suitable for a class of specially structured convex MOPs. Because multiobjective quadratic programs are a prominent class of convex MOPs due to their broad applicability, we review the state‐of‐the‐art algorithms for computing their efficient solutions. These two lines of investigation are merged to solve multiobjective portfolio optimization problems with three or more quadratic objective functions, a class of problems that has not been solved before. Computational examples are provided.

Document Details

Document Type
Pub Defense Publication
Publication Date
Nov 07, 2019
Source ID
10.1002/mcda.1690

Entities

People

  • Margaret M. Wiecek
  • Nathan Adelgren
  • Pubudu L. W. Jayasekara

Organizations

  • Clemson University
  • Office of Naval Research
  • PennWest Edinboro

Tags

Readers

  • Operations Research
  • Systems Analysis and Design