Clustering multivariate time series using energy distance
From MaRDI portal
characteristic functionclusteringdissimilarity measureenergy distancehierarchical clusteringstationaritytime series
Estimation in multivariate analysis (62H12) Measures of association (correlation, canonical correlation, etc.) (62H20) Classification and discrimination; cluster analysis (statistical aspects) (62H30) Time series, auto-correlation, regression, etc. in statistics (GARCH) (62M10) Inference from stochastic processes (62Mxx)
Abstract: A novel methodology is proposed for clustering multivariate time series data using energy distance defined in Sz'ekely and Rizzo (2013). Specifically, a dissimilarity matrix is formed using the energy distance statistic to measure separation between the finite dimensional distributions for the component time series. Once the pairwise dissimilarity matrix is calculated, a hierarchical clustering method is then applied to obtain the dendrogram. This procedure is completely nonparametric as the dissimilarities between stationary distributions are directly calculated without making any model assumptions. In order to justify this procedure, asymptotic properties of the energy distance estimates are derived for general stationary and ergodic time series. The method is illustrated in a simulation study for various component time series that are either linear or nonlinear. Finally the methodology is applied to two examples; one involves GDP of selected countries and the other is population size of various states in the U.S.A. in the years 1900 -1999.
Cites work
- A DISTANCE MEASURE FOR CLASSIFYING ARIMA MODELS
- A kernel two-sample test
- A periodogram-based metric for time series classification
- Applications of distance correlation to time series
- Asymptotic theory of statistical inference for time series
- Clustering of biological time series by cepstral coefficients based distances
- Clustering of time series data -- a survey
- Clusters of time series
- Comparing several parametric and nonparametric approaches to time series clustering: a simulation study
- Discrimination and Clustering for Multivariate Time Series
- Energy statistics: a class of statistics based on distances
- Ergodic theorems. With a supplement by Antoine Brunel
- scientific article; zbMATH DE number 4205743 (Why is no real title available?)
- scientific article; zbMATH DE number 3903784 (Why is no real title available?)
- scientific article; zbMATH DE number 3502497 (Why is no real title available?)
- Measuring and testing dependence by correlation of distances
- Mixing: Properties and examples
- Multivariate analysis in vector time series.
- Nonlinear time series clustering based on Kolmogorov-Smirnov 2D statistic
- Strong laws for 𝐿- and 𝑢-statistics
- Testing independence for multivariate time series via the auto-distance correlation matrix
- Time series clustering and classification
- Ward's hierarchical agglomerative clustering method: which algorithms implement Ward's criterion?
Cited in
(5)- MODIS NDVI time series clustering under dynamic time warping
- Clustering time series under the Fréchet distance
- Zero-inflated time series clustering via ensemble thick-pen transform
- Comonotonic-based time series clustering with constraints: a review and a conceptual framework
- Mathematics, statistics, and geometry of extreme events in high dimensions. Abstracts from the workshop held August 19--23, 2024
This page was built for publication: Clustering multivariate time series using energy distance
Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q6135361)
