Feature clustering with self-organizing maps and an application to financial time-series for portfolio selection

Bruno Silva, Nuno Marques

Research output: Chapter in Book/Report/Conference proceedingConference contributionpeer-review

20 Citations (Scopus)

Abstract

The portfolio selection is an important technique for decreasing the risk in the stock investment. In the portfolio selection, the investor's property is distributed for a set of stocks in order to minimize the financial risk in market downturns. With this in mind, and aiming to develop a tool to assist the investor in finding balanced portoflios, we achieved a generic method for feature clustering with Self-Organizing Maps (SOM). The ability of neural networks to discover nonlinear relationships in input data makes them ideal for modeling dynamic systems as the stock market. The method proposed makes use the remarkable visualization capabilities of the SOM, namely the Component Planes, to detect non-linear correlations between features. An appropriate metric - the improved Rv coefficient - is also proposed to compare Component Planes and generate a distance matrix between features, after which an hierarchical clustering method is used to obtain the clusters of features. Results obtained are empirically sound, although at this moment we do not provide mathematical comparisons with other methods. Results also show that feature clustering with the SOM presents itself as a viable method to cluster time-series.

Original languageEnglish
Title of host publicationICFC 2010 ICNC 2010 - Proceedings of the International Conference on Fuzzy Computation and International Conference on Neural Computation
Pages301-309
Number of pages9
Publication statusPublished - 2010
EventInternational Conference on Neural Computation, ICNC 2010 and of the International Conference on Fuzzy Computation, ICFC 2010 - Valencia, Spain
Duration: 24 Oct 201026 Oct 2010

Conference

ConferenceInternational Conference on Neural Computation, ICNC 2010 and of the International Conference on Fuzzy Computation, ICFC 2010
CountrySpain
CityValencia
Period24/10/1026/10/10

Keywords

  • Correlation hunting
  • Feature clustering
  • Portfolio selection
  • Self-organizing maps
  • Time-series clustering

Fingerprint

Dive into the research topics of 'Feature clustering with self-organizing maps and an application to financial time-series for portfolio selection'. Together they form a unique fingerprint.

Cite this