Please use this identifier to cite or link to this item:
Title: A review of portfolio planning: Models and systems
Authors: Mitra, G
Kyriakis, T
Lucas, CA
Pirbhai, M
Issue Date: 2003
Publisher: The Centre for the Analysis of Risk and Optimisation Modelling Applications (CARISMA), Brunel University
Citation: The Centre for the Analysis of Risk and Optimisation Modelling Applications (CARISMA), Brunel University; Technical Reports, CTR/01/03: Also appears in Advances in Portfolio Construction and Implementation. Satchell, S E and Scowcroft A E (Eds.) Butterworth and Heinemann, Oxford
Abstract: In this chapter, we first provide an overview of a number of portfolio planning models which have been proposed and investigated over the last forty years. We revisit the mean-variance (M-V) model of Markowitz and the construction of the risk-return efficient frontier. A piecewise linear approximation of the problem through a reformulation involving diagonalisation of the quadratic form into a variable separable function is also considered. A few other models, such as, the Mean Absolute Deviation (MAD), the Weighted Goal Programming (WGP) and the Minimax (MM) model which use alternative metrics for risk are also introduced, compared and contrasted. Recently asymmetric measures of risk have gained in importance; we consider a generic representation and a number of alternative symmetric and asymmetric measures of risk which find use in the evaluation of portfolios. There are a number of modelling and computational considerations which have been introduced into practical portfolio planning problems. These include: (a) buy-in thresholds for assets, (b) restriction on the number of assets (cardinality constraints), (c) transaction roundlot restrictions. Practical portfolio models may also include (d) dedication of cashflow streams, and, (e) immunization which involves duration matching and convexity constraints. The modelling issues in respect of these features are discussed. Many of these features lead to discrete restrictions involving zero-one and general integer variables which make the resulting model a quadratic mixed-integer programming model (QMIP). The QMIP is a NP-hard problem; the algorithms and solution methods for this class of problems are also discussed. The issues of preparing the analytic data (financial datamarts) for this family of portfolio planning problems are examined. We finally present computational results which provide some indication of the state-of-the-art in the solution of portfolio optimisation problems.
Appears in Collections:Dept of Mathematics Research Papers
Mathematical Sciences

Files in This Item:
File Description SizeFormat 
CTR-01-03%20T%20Kyriakis.pdf797.26 kBAdobe PDFView/Open

Items in BURA are protected by copyright, with all rights reserved, unless otherwise indicated.