Greedy Gaussian segmentation of multivariate time series
From MaRDI portal
Publication:2324258
Abstract: We consider the problem of breaking a multivariate (vector) time series into segments over which the data is well explained as independent samples from a Gaussian distribution. We formulate this as a covariance-regularized maximum likelihood problem, which can be reduced to a combinatorial optimization problem of searching over the possible breakpoints, or segment boundaries. This problem can be solved using dynamic programming, with complexity that grows with the square of the time series length. We propose a heuristic method that approximately solves the problem in linear time with respect to this length, and always yields a locally optimal choice, in the sense that no change of any one breakpoint improves the objective. Our method, which we call greedy Gaussian segmentation (GGS), easily scales to problems with vectors of dimension over 1000 and time series of arbitrary length. We discuss methods that can be used to validate such a model using data, and also to automatically choose appropriate values of the two hyperparameters in the method. Finally, we illustrate our GGS approach on financial time series and Wikipedia text data.
Recommendations
- Time-series segmentation: A model and a method
- An L₀-norm regularized method for multivariate time series segmentation
- Joint segmentation of multivariate Gaussian processes using mixed linear models
- A Total Variation Based Method for Multivariate Time Series Segmentation
- Modified Gath-Geva clustering for fuzzy segmentation of multivariate time-series
Cites work
- $\ell_1$ Trend Filtering
- A Bayesian approach to retrospective identification of change-points
- A dynamic programming algorithm for linear text segmentation
- A dynamic programming segmentation procedure for hydrological and environmental time series
- A well-conditioned estimator for large-dimensional covariance matrices
- Bayesian analysis of a change-point in exponential families with applications.
- Bayesian single change point detection in a sequence of multivariate normal observations
- Covariance matrix selection and estimation via penalised normal likelihood
- Covariance-regularized regression and classification for high dimensional problems
- Detecting change-points in multidimensional stochastic processes
- Detection of abrupt changes: theory and application
- Efficient Greedy Learning of Gaussian Mixture Models
- Enhancing sparsity by reweighted \(\ell _{1}\) minimization
- Finding multiple abrupt change points
- scientific article; zbMATH DE number 3809326 (Why is no real title available?)
- Inference in hidden Markov models.
- Joint segmentation of multivariate Gaussian processes using mixed linear models
- Long memory of financial time series and hidden Markov models with time-varying parameters
- Model-based clustering and segmentation of time series with changes in regime
- Modified Gath-Geva clustering for fuzzy segmentation of multivariate time-series
- Multiple break detection in the correlation structure of random variables
- Multiple change-point detection of multivariate mean vectors with the Bayesian approach
- Multivariate Generalizations of Cumulative Sum Quality-Control Schemes
- Nonparametric multivariate statistical process control charts: a hypothesis testing-based approach
- On marginal likelihood computation in change-point models
- On the approximation of curves by line segments using dynamic programming
- Regularized estimation of large covariance matrices
- Sparsity and Smoothness Via the Fused Lasso
- Symbolic time series analysis via wavelet-based partitioning
- Time-series data mining
- Using the minimum description length to discover the intrinsic cardinality and dimensionality of time series
Cited in
(12)- Time-series segmentation: A model and a method
- Segmental dynamic factor analysis for time series of curves
- Sparse group fused Lasso for model segmentation: a hybrid approach
- Joint segmentation of multivariate Gaussian processes using mixed linear models
- Multiscale and multilevel technique for consistent segmentation of nonstationary time series
- An L₀-norm regularized method for multivariate time series segmentation
- Change-Point Detection for Graphical Models in the Presence of Missing Values
- Forecasting market states
- A Total Variation Based Method for Multivariate Time Series Segmentation
- Robust multiscale estimation of time-average variance for time series segmentation
- A temporal proximity regularized clusterwise vector autoregressive method for multivariate time series clustering
- Segmentation, classification and denoising of a time series field by a variational method
This page was built for publication: Greedy Gaussian segmentation of multivariate time series
Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q2324258)
