Clustering of financial time series in extreme scenarios
Roberta Pappadà, Fabrizio Durante · Open Research (University of Surrey) · 2012
A methodology is presented for clustering financial time series in extreme scenario. The procedure is based on the calculation of some suitable pairwise conditional Spearman’s correlation coefficients. It does not assume any parametric model describing the time series under investigation, but only relies on the assumption that they follows a multivariate copula-GARCH model.