Correlation based hierarchical clustering in financial time series
- Autori: Micciche', S.; Lillo, F.; Mantegna, R.
- Anno di pubblicazione: 2005
- Tipologia: Proceedings (TIPOLOGIA NON ATTIVA)
We review a correlation based clustering procedure applied to a portfolio of assets synchronously traded in a financial market. The portfolio considered consists of the set of 500 highly capitalized stocks traded at the New York Stock Exchange during the time period 1987-1998. We show that meaningful economic information can be extracted from correlation matrices.