Data clustering and noise undressing for correlation matrices
arXiv:cond-mat/0101237 · doi:10.1103/PhysRevE.63.061101
Abstract
We discuss a new approach to data clustering. We find that maximum likelihood leads naturally to an Hamiltonian of Potts variables which depends on the correlation matrix and whose low temperature behavior describes the correlation structure of the data. For random, uncorrelated data sets no correlation structure emerges. On the other hand for data sets with a built-in cluster structure, the method is able to detect and recover efficiently that structure. Finally we apply the method to financial time series, where the low temperature behavior reveals a non trivial clustering.
8 pages, 5 figures, completely rewritten and enlarged version of cond-mat/0003241. Submitted to Phys. Rev. E
Cited by in corpus (29)
- Intensity and coherence of motifs in weighted complex networks
- Networks of equities in financial markets
- Correlation, hierarchies, and networks in financial markets
- Correlation based networks of equity returns sampled at different time horizons
- Random Matrix Theory Analysis of Cross Correlations in Financial Markets
- Collective behavior of stock price movements in an emerging market
- Systematic analysis of group identification in stock markets
- Spanning Trees and bootstrap reliability estimation in correlation based networks
- Emergence of time-horizon invariant correlation structure in financial returns by subtraction of the market mode
- A New Method to Estimate the Noise in Financial Correlation Matrices
- An analysis of Cross-correlations in South African Market data
- Collective Origin of the Coexistence of Apparent RMT Noise and Factors in Large Sample Correlation Matrices
- Hierarchically nested factor model from multivariate data
- Algorithms of maximum likelihood data clustering with applications
- Maximal spanning trees, asset graphs and random matrix denoising in the analysis of dynamics of financial networks
- Credit risk - A structural model with jumps and correlations
- Quantifying Relevance in Learning and Inference
- Cross-correlation asymmetries and causal relationships between stock and market risk
- Detecting intraday financial market states using temporal clustering
- Spectral methods and cluster structure in correlation-based networks
- Bootstrap validation of links of a minimum spanning tree
- Hierarchical causality in financial economics
- Macrostate Data Clustering
- High-speed detection of emergent market clustering via an unsupervised parallel genetic algorithm
- A Quantitative Clustering Approach to Ultrametricity in Spin Glasses
- Fast Super-Paramagnetic Clustering
- Inferring financial stock returns correlation from complex network analysis
- Agglomerative Likelihood Clustering
- Random Matrix Application to Correlations Among Volatility of Assets