Economic sector identification in a set of stocks traded at the New York Stock Exchange: a comparative analysis

Abstract

We review some methods recently used in the literature to detect the existence of a certain degree of common behavior of stock returns belonging to the same economic sector. Specifically, we discuss methods based on random matrix theory and hierarchical clustering techniques. We apply these methods to a set of stocks traded at the New York Stock Exchange. The investigated time series are recorded at a daily time horizon.All the considered methods are able to detect economic information and the presence of clusters characterized by the economic sector of stocks. However, different methodologies provide different information about the considered set. Our comparative analysis suggests that the application of just a single method could not be able to extract all the economic information present in the correlation coefficient matrix of a set of stocks.

Spanning Trees and Bootstrap Reliability Estimation in Correlation-Based Networks

“…Attempts to overcome these problems may require to filter out statistically reliable information from the correlation matrix. Classical spectral methods [9] and related refinements [6,7], clustering [10] and graph theory [11] revealed to be fruitful approaches to the analysis of empirical data of correlation based systems [3,8,12,13,14,15,16,17,18,19,20,21,22,23,24]. In this paper we exploit the deep relation between the Single Linkage Cluster Analysis (SLCA) [10] which is a classical technique of hierarchical clustering, and the Minimum Spanning Tree (MST) [11] which is a classical concept of graph theory [25], to introduce a new tree that we name Average Linkage Minimum Spanning Tree (ALMST).…”

“…This method is widely used in phylogenetic analysis where it is known as "unweighted pair group method using arithmetic averages" (UPGMA) [26]. The motivation for introducing the ALMST as a counterpart in graph theory of ALCA bears both on the relevance of the ALCA in the study of correlation based systems such as gene expression data [3,12,13,14] and stock returns [23] and on the extra information carried by the MST (ALMST) with respect to the SLCA (ALCA) 1 . In our investigation we have observed that the MST and ALMST are sharing information usually accounted by a significant percentage of common links.…”

Emergence of time-horizon invariant correlation structure in financial returns by subtraction of the market mode

“…Indeed, the HT gives a description of the hierarchical organization of the stocks, while the MST gives an indication about their topological organization. For a review of SLCA in the context of multivariate financial time series we refer to [13,20,21].…”

“…Interestingly, it was found that such a process is much faster today than in the past and more pronounced for more capitalized stocks [9]. It has also been remarked [12,13] that the structure of correlations changes as the time-horizon τ over which returns are defined increases, i.e. that "pictorially, the market appears as an embryo which progressively forms and differentiates over time" [10].…”

Correlations and Dependencies in the Global Financial Village

“…Several methods have been developed to overcome this obstacle by constructing from the complete network a sub-graph that captures the most relevant information embedded in the original network. A widely used method to construct informative sub-graph of a complete network is the Minimum Spanning Tree (MST) 24,25,26,27,28,29 . Another informative sub-graph which retains more information (in comparison to the MST) is the Planar Maximally Filtered Graph (PMFG) 20,30 .…”