|
Descripción
|
|
|---|---|
| The modern theory of portfolio optimization emerged with the mean-variance model, dating back to the papers of Roy and Markowitz. Because of the widespread acceptance of Markowitz's approach, his name is virtually synonymous with portfolio selection. Markowitz essentially computes the entire non-dominated set first and, then, the investor attempt to identify a most preferred portfolio. More recently, the classical Markowitz model has been extended in different ways, like the inclusion of cardinality constraints that limit a portfolio to have a specified number of assets, and to impose limits on the proportion of the portfolio held in a given asset, or the construction of decision tables in the second phase of Markowitz model by considering multiple scenarios assuming strict uncertainty. In recent years, criticism of the basic model has been increasing because of its disregard for individual investors preferences. Investors often prefer portfolios that lie behind the non-dominated frontier of the Markowitz model even though they are dominated by other portfolios. A possible explanation is that not all the relevant information for an investment decision can be captured in terms of explicit return and risk. Consequently, a multicriteria model based on more than two objective functions allows for a higher flexibility in modeling the objectives of investors, and, combined, for instance, with an appropriate utility approach, is likely to lead to better representations of their preferences. Within the multi-attribute or multicriteria approach, several forms of aggregation models have been developed to support the portfolio selection, like Outranking Relations, Multi-Attribute Utility Theory (MAUT), Analytic Hierarchy Process (AHP), Goal and Compromise Programming, Preference disaggregation, Preference Programming and Rough Set Theory. We provide an extension of the classical portfolio selection approach, utilizing a decision support system based on an additive multi-attribute utility model. The 27 3rd International Workshop on Multi-attribute Methods in Finance and Insurance approach considers the portfolio problem with several objectives (return, risk and liquidity) simultaneously to better reflect the investor's preferences and allows for imprecise information about them, which are elicited by means of the GMAA system. Generic Multi-Attribute Analysis System (http://www.dia.fi.upm.es/¿ajimenez/GMAA) is a PC-based decision support system based on an additive multi-attribute utility model that is intended to allay many of the operational difficulties involved in the decision analysis cycle. It admits incomplete information about the investor¿s preferences through value intervals as responses to the probability questions the investors are asked. Thus it can process incomplete preference statements and provides facilities to assess imprecise utility functions and scaling constants. The resulting optimization problem is difficult to solve due to the complexity of the objective function, which is has many local maxima in reduced neighborhoods. Consequently, we decided to compute the global maximum using an algorithm based on an evolutionary computation, which has been proven by different authors that provides better results to tackle with different portfolio optimization models regarding other metaheuristics. Finally, the methodology is illustrated considering an application to Madrid Stock Market securities of the Ibex 35 for the latest two consecutive years. | |
|
Internacional
|
Si |
|
Nombre congreso
|
3rd International Workshop on Multi-attribute Methods in Finance and Insurance |
|
Tipo de participación
|
960 |
|
Lugar del congreso
|
Madrid |
|
Revisores
|
Si |
|
ISBN o ISSN
|
978-3-642-01019-4 |
|
DOI
|
|
|
Fecha inicio congreso
|
22/04/2009 |
|
Fecha fin congreso
|
24/04/2009 |
|
Desde la página
|
27 |
|
Hasta la página
|
28 |
|
Título de las actas
|
Proceedings of the 3rd International Workshop on Multi-attribute Methods in Finance and Insurance |