full
search.png

Search

close.png

Cancel

arrow
Volume 12, Issue 2
An $L_0$-Norm Regularized Method for Multivariate Time Series Segmentation

Min Li & Yu-Mei Huang

DOI: 10.4208/eajam.180921.050122

East Asian J. Appl. Math., 12 (2022), pp. 353-366.

Published online: 2022-02

Preview Purchase PDF 2504 40135

Cited by

google scholar semantic scholar
Export citation
  • Abstract

A multivariate time series segmentation model based on the minimization of the negative log-likelihood function of the series is proposed. The model is regularized by the $L_0$-norm of the time series mean change and solved by an alternating process. We use a dynamic programming algorithm in order to determine the breakpoints and the cross-validation method to find the parameters of the model. Experiments show the efficiency of the method for segmenting both synthetic and real multivariate time series.

  • Keywords

Multivariate time series, segmentation, $L_0$-norm, dynamic programming.

  • AMS Subject Headings

37M10, 62M10, 91B84

  • Copyright

COPYRIGHT: © Global Science Press

  • Email address
  • BibTex
  • RIS
  • TXT
@Article{EAJAM-12-353, author = {Li , Min and Huang , Yu-Mei}, title = {An $L_0$-Norm Regularized Method for Multivariate Time Series Segmentation}, journal = {East Asian Journal on Applied Mathematics}, year = {2022}, volume = {12}, number = {2}, pages = {353--366}, abstract = {

A multivariate time series segmentation model based on the minimization of the negative log-likelihood function of the series is proposed. The model is regularized by the $L_0$-norm of the time series mean change and solved by an alternating process. We use a dynamic programming algorithm in order to determine the breakpoints and the cross-validation method to find the parameters of the model. Experiments show the efficiency of the method for segmenting both synthetic and real multivariate time series.

}, issn = {2079-7370}, doi = {https://doi.org/10.4208/eajam.180921.050122}, url = {http://global-sci.org/intro/article_detail/eajam/20258.html} }
TY - JOUR T1 - An $L_0$-Norm Regularized Method for Multivariate Time Series Segmentation AU - Li , Min AU - Huang , Yu-Mei JO - East Asian Journal on Applied Mathematics VL - 2 SP - 353 EP - 366 PY - 2022 DA - 2022/02 SN - 12 DO - http://doi.org/10.4208/eajam.180921.050122 UR - https://global-sci.org/intro/article_detail/eajam/20258.html KW - Multivariate time series, segmentation, $L_0$-norm, dynamic programming. AB -

A multivariate time series segmentation model based on the minimization of the negative log-likelihood function of the series is proposed. The model is regularized by the $L_0$-norm of the time series mean change and solved by an alternating process. We use a dynamic programming algorithm in order to determine the breakpoints and the cross-validation method to find the parameters of the model. Experiments show the efficiency of the method for segmenting both synthetic and real multivariate time series.

Li , Min and Huang , Yu-Mei. (2022). An $L_0$-Norm Regularized Method for Multivariate Time Series Segmentation. East Asian Journal on Applied Mathematics. 12 (2). 353-366. doi:10.4208/eajam.180921.050122
Copy to clipboard
BibteX RIS TXT
The citation has been copied to your clipboard