Community detection for correlation matrices

Abstract

A challenging problem in the study of complex systems is that of resolving, without prior information, the emergent, mesoscopic organization determined by groups of units whose dynamical activity is more strongly correlated internally than with the rest of the system. The existing techniques to filter correlations are not explicitly oriented towards identifying such modules and can suffer from an unavoidable information loss. A promising alternative is that of employing community detection techniques developed in network theory. Unfortunately, this approach has focused predominantly on replacing network data with correlation matrices, a procedure that tends to be intrinsically biased due to its inconsistency with the null hypotheses underlying the existing algorithms. Here we introduce, via a consistent redefinition of null models based on random matrix theory, the appropriate correlation-based counterparts of the most popular community detection techniques. Our methods can filter out both unit-specific noise and system-wide dependencies, and the resulting communities are internally correlated and mutually anti-correlated. We also implement multiresolution and multifrequency approaches revealing hierarchically nested sub-communities with `hard' cores and `soft' peripheries. We apply our techniques to several financial time series and identify mesoscopic groups of stocks which are irreducible to a standard, sectorial taxonomy, detect `soft stocks' that alternate between communities, and discuss implications for portfolio optimization and risk management.

Final version, accepted for publication on PRX

References in corpus (14)

• Fast unfolding of communities in large networks
• Modularity and community structure in networks
• Resolution limit in community detection
• Statistical Mechanics of Community Detection
• A tool for filtering information in complex systems
• Analysis of the structure of complex networks at different resolution levels
• Analysis of community structure in networks of correlated data
• Maximum likelihood: extracting unbiased information from complex networks
• Collective behavior of stock price movements in an emerging market
• Generalized Bose-Fermi statistics and structural correlations in weighted networks
• Kullback-Leibler distance as a measure of the information filtered from multivariate data
• Multi-scale Modularity in Complex Networks
• Maximal spanning trees, asset graphs and random matrix denoising in the analysis of dynamics of financial networks
• Detecting modules in dense weighted networks with the Potts method

Cited by in corpus (41)

• Dynamic fluctuations coincide with periods of high and low modularity in resting-state functional brain networks
• The Physics of Financial Networks
• Inferring the mesoscale structure of layered, edge-valued and time-varying networks
• Modeling sequences and temporal networks with dynamic community structures
• Nonparametric weighted stochastic block models
• Stock market as temporal network
• Finding multiple core-periphery pairs in networks
• Core-periphery structure requires something else in the network
• Vulnerability and co-susceptibility determine the size of network cascades
• Descriptive vs. inferential community detection in networks: pitfalls, myths, and half-truths
• Multiscale dynamical embeddings of complex networks
• Hierarchical organization of functional connectivity in the mouse brain: a complex network approach
• Uncovering the Dynamics of Correlation Structures Relative to the Collective Market Motion
• Mesoscopic Community Structure of Financial Markets Revealed by Price and Sign Fluctuations
• Configuration model for correlation matrices preserving the node strength
• Functional brain modules reconfigure at multiple scales across the human lifespan
• Dynamics of Investor Spanning Trees Around Dot-Com Bubble
• Meta-validation of bipartite network projections
• Introduction to correlation networks: Interdisciplinary approaches beyond thresholding
• Hierarchical mutual information for the comparison of hierarchical community structures in complex networks
• Modularity and the spread of perturbations in complex dynamical systems
• Gene communities in co-expression networks across different tissues
• Uncovering functional brain signature via random matrix theory
• A New Attempt to Identify Long-term Precursors for Endogenous Financial Crises in the Market Correlation Structures
• The specificity and robustness of long-distance connections in weighted, interareal connectomes
• Correlations and Clustering in Wholesale Electricity Markets
• Random matrix approach to multivariate categorical data analysis
• Multi-scale Laplacian community detection in heterogeneous networks
• Decomposition Pipeline for Large-Scale Portfolio Optimization with Applications to Near-Term Quantum Computing
• Integrated Information Decomposition Unveils Major Structural Traits of $In$ $Silico$ and $In$ $Vitro$ Neuronal Networks
• Multiple Partitioning of Multiplex Signed Networks: Application to European Parliament Votes
• Multiplicity and Diversity: Analyzing the Optimal Solution Space of the Correlation Clustering Problem on Complete Signed Graphs
• Whole-Graph Representation Learning For the Classification of Signed Networks
• Quantum generative modeling for financial time series with temporal correlations
• The Importance of Being Negative: A serious treatment of non-trivial edges in brain functional connectome
• Skeleton coupling: a novel interlayer mapping of community evolution in temporal networks
• Correlation-Based Community Detection
• A random matrix perspective of cultural structure: groups or redundancies?
• Assessing imbalance in signed brain networks
• Detectability threshold in weighted modular networks
• The interdependency structure in the Mexican stock exchange: A network approach