Open Access
2019 Matrix factorization for multivariate time series analysis
Pierre Alquier, Nicolas Marie
Electron. J. Statist. 13(2): 4346-4366 (2019). DOI: 10.1214/19-EJS1630


Matrix factorization is a powerful data analysis tool. It has been used in multivariate time series analysis, leading to the decomposition of the series in a small set of latent factors. However, little is known on the statistical performances of matrix factorization for time series. In this paper, we extend the results known for matrix estimation in the i.i.d setting to time series. Moreover, we prove that when the series exhibit some additional structure like periodicity or smoothness, it is possible to improve on the classical rates of convergence.


Download Citation

Pierre Alquier. Nicolas Marie. "Matrix factorization for multivariate time series analysis." Electron. J. Statist. 13 (2) 4346 - 4366, 2019.


Received: 1 March 2019; Published: 2019
First available in Project Euclid: 6 November 2019

zbMATH: 07136618
MathSciNet: MR4028508
Digital Object Identifier: 10.1214/19-EJS1630

Primary: 62M20
Secondary: 60B20 , 60G35 , 62G08 , 62H12 , 62H25 , 62M10 , 93E14

Keywords: matrix factorization , Multivariate Time Series Analysis , Non-parametric regression , random matrices

Vol.13 • No. 2 • 2019
Back to Top