Taxonomy of Stock Market Indices

Preprint
Publisher Website
Google Scholar
  • 19 January 2000
Abstract
We investigate a set of time series of financial non-redundant time series. The set is composed by 29 stock market indices located all over the world in five continents. The correlation matrix is used to determine a distance matrix between the stock indices. By using the ultrametric matrix associated with the distance matrix we determine a meaningful taxonomy of the investigated stock indices. The detection of such a taxonomy proves that interpretable information can be stored in unpredictable non-redundant time series.

This publication has 0 references indexed in Scilit: