Relation between financial market structure and the real economy: Comparison between clustering methods

Abstract

We quantify the amount of information ltered by dierent hierarchical clustering methods on correlations between stock returns comparing the clustering structure with the underlying industrial activity classication. We apply, for the rst time to nancial data, a novel hierarchical clustering approach, the Directed Bubble Hierarchical Tree and we compare it with other methods including the Linkage and k-medoids. By taking the industrial sector classication of stocks as a benchmark partition, we evaluate how the dierent methods retrieve this classication. The results show that the Directed Bubble Hierarchical Tree can outperform other methods, being able to retrieve more information with fewer clusters. Moreover, we show that the economic information is hidden at dierent levels of the hierarchical structures depending on the clustering method. The dynamical analysis on a rolling window also reveals that the dierent methods show dierent degrees of sensitivity to events aecting nancial markets, like crises. These results can be of interest for all the applications of clustering methods to portfolio optimization and risk hedging.