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

Abstract

The evolution with time of the correlation structure of equity returns is studied by means of a filtered network approach investigating persistences and recurrences and their implications for risk diversification strategies. We build dynamically Planar Maximally Filtered Graphs from the correlation structure over a rolling window and we study the persistence of the associated Directed Bubble Hierarchical Tree (DBHT) clustering structure. We observe that the DBHT clustering structure is quite stable during the early 2000' becoming gradually less persistent before the unfolding of the 2007-2008 crisis. The correlation structure eventually recovers persistence in the aftermath of the crisis settling up a new phase, distinct from the pre-cysts structure, where the market structure is less related to industrial sector activity. Notably, we observe that -presently-the correlation structure is loosing again persistence indicating the building-up of another, different, phase. Such dynamical changes in persistence and their occurrence at the unfolding of financial crises rises concerns about the effectiveness of correlation-based portfolio management tools for risk diversification.

Parsimonious modeling with information filtering networks

“…Once the sparse precision matrix has been estimated, a number of efficient tools-mostly based on research in sparse numerical linear algebra-can be used to sample from the distribution, calculate conditional probabilities, calculate conditional statistics, and forecast [2,3]. GMRFs are of great importance in many applications spanning computer vision [4], sparse sensing [5], finance [6][7][8][9][10][11], gene expression [12][13][14]; biological neural networks [15], climate networks [16,17]; geostatistics and spatial statistics [18][19][20]. Almost universally, applications require modeling a large number of variables with a relatively small number of observations, and therefore the issue of the statistical significance of the model parameters is very important.…”

“…The novelty of the method presented in this paper is the combination of decomposable information filtering networks [6][7][8][9][10] with Gaussian Markov random fields [1,2] to produce parsimonious models associated with a meaningful structure of dependency between the variables. The strength of this methodology is that the global sparse inverse covariance matrix is produced from a simple sum of local inversions.…”

Stability and hierarchy of quasi-stationary states: financial markets as an example

“…The self-organization may produce patterns in observed data which are generally difficult to uncover. A wide range of data analysis techniques is available and widely used, including graph theoretical information filtering [2,3,4,5,6,7], data clustering [8,9,10,11,12,13] and geometric approaches [14,15,16,17,18]. All these techniques are based on a similarity measure between the data points.…”

The Multiplex Dependency Structure of Financial Markets

“…In general, each of the four layers is contributing information that cannot be found in the other three layers. It was shown in a recent paper by some of the authors [36] that information filtering networks can be used to forecast volatility outbursts. The present results suggest that a multilayer approach could provide a further forecasting instrument for bear/bull markets.…”