Evolution of correlation structure of industrial indices of U.S. equity markets

Abstract

We investigate the dynamics of correlations present between pairs of industry indices of US stocks traded in US markets by studying correlation based networks and spectral properties of the correlation matrix. The study is performed by using 49 industry index time series computed by K. French and E. Fama during the time period from July 1969 to December 2011 that is spanning more than 40 years. We show that the correlation between industry indices presents both a fast and a slow dynamics. The slow dynamics has a time scale longer than five years showing that a different degree of diversification of the investment is possible in different periods of time. On top to this slow dynamics, we also detect a fast dynamics associated with exogenous or endogenous events. The fast time scale we use is a monthly time scale and the evaluation time period is a 3 month time period. By investigating the correlation dynamics monthly, we are able to detect two examples of fast variations in the first and second eigenvalue of the correlation matrix. The first occurs during the dot-com bubble (from March 1999 to April 2001) and the second occurs during the period of highest impact of the subprime crisis

Risk diversification: a study of persistence with a filtered correlation-network approach

“…For instance, the Financial sector appears to loose centrality over the first decade of 2000's . In Buccheri et al (2013) the authors found both a slow and a fast dynamics in correlation networks topology: while the slow dynamics shows persistence over periods of at least 5 years, the time scale of the fast dynamics is of order of few months and it is linked to special exogenous and endogenous events like financial crises. For instance, in Onnela et al (2003a) it has been shown that sharp structural changes occurred in the graph topology during the 1987 Black Monday.…”

Immediate causality network of stock markets

“…Based upon the top ten financial events each year reported in the Financial Times (in Chinese) [21] and the discussions in references [5,6,[9][10][11] of the present paper, eight crises are identified (see Table I). Let a window with predefined length and step slide along the return series.…”

“…Complex network theory has been adopted in some works to monitor evolutionary behaviors of financial systems [3][4][5][6][7][8][9][10][11][12][13][14][15][16][17][18]. For instance, Song et al [5] convert crosscorrelations between a total of 57 industry portfolios in the American stock market into a series of planar maximally filtered graphs (a network embedded in a highdimensional surface) to represent the states of the corresponding successive durations, and find that mutual entropy between successive states reaches a peak at a financial crisis.…”

Emergence of correlations between securities at short time scales

“…An empirical validation is performed on long time series of 1-min returns of a large universe of U.S. stocks. To get several significant eigenvalues at time scales from 1 minute to 2 hours, the correlation matrix was estimated over the whole available period (2013-2017) so that variations of cross-correlations over time were ignored (note that the dynamics of the eigenvalues and eigenvectors over time has been investigated elsewhere [16,17,18]). In spite of its simple character, the lead-lag factor model is shown to be able to reproduce the dependence of large eigenvalues on the time scale.…”