CFE 2020: Start Registration
View Submission - CMStatistics
B0693
Title: A novel structure-based approach for multivariate time series clustering Authors:  Angel Lopez Oriona - King Abdullah University of Science and Technology (KAUST) (Saudi Arabia) [presenting]
Jose Vilar - Universidade da Coruna (Spain)
Abstract: Clustering of multivariate time series (MTS) is a central problem in data mining with many applications. Frequently, the clustering target is to identify groups of MTS generated by the same multivariate stochastic process. Most of the approaches to address this problem include a prior step of dimensionality reduction which may result in a loss of information on the structural relationships of the MTS or consider dissimilarities based on correlations and cross-correlations, but ignoring the serial dependence structure. We propose a novel approach to measure dissimilarity between MTS aimed at jointly measuring both cross-sectional and serial dependence. Each MTS is characterized by a set of matrices of estimated quantile cross-spectral densities, where each matrix corresponds to an arbitrary pair of quantile levels. Then the dissimilarity between every couple of MTS is evaluated by comparing their estimated quantile cross-spectral densities, and the pairwise dissimilarity matrix is taken as a starting point to develop a partitioning around medoids (PAM) algorithm. Since the quantile-based cross-spectra capture dependence in quantiles of the joint distribution, the proposed metric has a high capability to discriminate between high-level dependence structures. An extensive simulation study shows that our clustering procedure outperforms a wide range of alternative methods, besides being computationally efficient. A real data application illustrates the usefulness of our approach.