Multi-Period Portfolio Optimization Model with Cone Constraints and Discrete Decisions

28 Pages Posted: 15 Mar 2017 Last revised: 8 Jan 2019

See all articles by Ümit Saglam

Ümit Saglam

East Tennessee State University

Hande Yurttan Benson

Drexel University

Date Written: January 1, 2019

Abstract

In this study, we consider multi-period portfolio optimization model that is formulated as a mixed-integer second-order cone programming problems (MISOCPs). The Markowitz (1952) mean/variance framework has been extended by including transaction costs, conditional value-at-risk (CVaR), diversification-by-sector and buy-in thresholds constraints. The model is obtained using a binary scenario tree that is constructed with monthly returns of the stocks from the S&P 500. We solve these models with a MATLAB based Mixed Integer Linear and Nonlinear Optimizer (MILANO). Numerical results show that we can solve small to medium-sized instances successfully, and we provide a substantial improvement in runtimes using warmstarts in outer approximation algorithm.

Keywords: Multi-Period Portfolio Optimization, Mixed-Integer Second-Order Cone Programming, Mixed Integer Linear and Nonlinear Optimizer (MILANO), Outer Approximation

JEL Classification: C61, C63, C88, G11

Suggested Citation

Saglam, Ümit and Benson, Hande Yurttan, Multi-Period Portfolio Optimization Model with Cone Constraints and Discrete Decisions (January 1, 2019). Available at SSRN: https://ssrn.com/abstract=2932567 or http://dx.doi.org/10.2139/ssrn.2932567

Ümit Saglam (Contact Author)

East Tennessee State University ( email )

Department of Management and Marketing
PO Box 70625
Johnson City, TN 37614
United States
423-439-1000 (Phone)
423-439-4422 (Fax)

HOME PAGE: http://https://sites.google.com/view/umitsaglam/home

Hande Yurttan Benson

Drexel University ( email )

3141 Chestnut St
Philadelphia, PA 19104
United States

Register to save articles to
your library

Register

Paper statistics

Downloads
191
rank
154,914
Abstract Views
557
PlumX Metrics