Factor Models for High-Dimensional Tensor Time Series

Rong Chen; Dan Yang; Cun Hui Zhang

doi:https://doi.org/10.1080/01621459.2021.1912757

Factor Models for High-Dimensional Tensor Time Series

Rong Chen, Dan Yang, Cun Hui Zhang

School of Arts and Sciences, Statistics

Rutgers, The State University

Research output: Contribution to journal › Article › peer-review

Abstract

Large tensor (multi-dimensional array) data routinely appear nowadays in a wide range of applications, due to modern data collection capabilities. Often such observations are taken over time, forming tensor time series. In this article we present a factor model approach to the analysis of high-dimensional dynamic tensor time series and multi-category dynamic transport networks. This article presents two estimation procedures along with their theoretical properties and simulation results. We present two applications to illustrate the model and its interpretations.

Original language	American English
Pages (from-to)	94-116
Number of pages	23
Journal	Journal of the American Statistical Association
Volume	117
Issue number	537
DOIs	https://doi.org/10.1080/01621459.2021.1912757
State	Published - 2022

ASJC Scopus subject areas

Statistics and Probability
Statistics, Probability and Uncertainty

Keywords

Autocovariance matrices
Cross-covariance matrices
Dimension reduction
Dynamic transport network
Eigen-analysis
Factor models
Import–export
Tensor time series
Traffic
Unfolding

Access to Document

https://doi.org/10.1080/01621459.2021.1912757

Cite this

@article{83e358af92734017b42f72101e147b3a,

title = "Factor Models for High-Dimensional Tensor Time Series",

abstract = "Large tensor (multi-dimensional array) data routinely appear nowadays in a wide range of applications, due to modern data collection capabilities. Often such observations are taken over time, forming tensor time series. In this article we present a factor model approach to the analysis of high-dimensional dynamic tensor time series and multi-category dynamic transport networks. This article presents two estimation procedures along with their theoretical properties and simulation results. We present two applications to illustrate the model and its interpretations.",

keywords = "Autocovariance matrices, Cross-covariance matrices, Dimension reduction, Dynamic transport network, Eigen-analysis, Factor models, Import–export, Tensor time series, Traffic, Unfolding",

author = "Rong Chen and Dan Yang and Zhang, {Cun Hui}",

note = "Funding Information: Chen{\textquoteright}s research is supported in part by National Science Foundation grants DMS-1503409, DMS-1737857, IIS-1741390, CCF-1934924 and DMS-2027855. Yang{\textquoteright}s research is supported in part by NSF grant IIS-1741390, Hong Kong GRF 17301620 and CRF C7162-20GF. Zhang{\textquoteright}s research is supported in part by NSF grants DMS-1721495, IIS-1741390 and CCF-1934924. The authors thank the editor, an associate editor, and three anonymous referees for their helpful comments. Division of Computing and Communication Foundations; Division of Information and Intelligent Systems;Hong Kong CRF;Division of Mathematical Sciences;Division of Mathematical Sciences; Publisher Copyright: {\textcopyright} 2021 American Statistical Association.",

year = "2022",

doi = "https://doi.org/10.1080/01621459.2021.1912757",

language = "American English",

volume = "117",

pages = "94--116",

journal = "Journal of the American Statistical Association",

issn = "0162-1459",

publisher = "Taylor and Francis Ltd.",

number = "537",

}

TY - JOUR

T1 - Factor Models for High-Dimensional Tensor Time Series

AU - Chen, Rong

AU - Yang, Dan

AU - Zhang, Cun Hui

N1 - Funding Information: Chen’s research is supported in part by National Science Foundation grants DMS-1503409, DMS-1737857, IIS-1741390, CCF-1934924 and DMS-2027855. Yang’s research is supported in part by NSF grant IIS-1741390, Hong Kong GRF 17301620 and CRF C7162-20GF. Zhang’s research is supported in part by NSF grants DMS-1721495, IIS-1741390 and CCF-1934924. The authors thank the editor, an associate editor, and three anonymous referees for their helpful comments. Division of Computing and Communication Foundations; Division of Information and Intelligent Systems;Hong Kong CRF;Division of Mathematical Sciences;Division of Mathematical Sciences; Publisher Copyright: © 2021 American Statistical Association.

PY - 2022

Y1 - 2022

N2 - Large tensor (multi-dimensional array) data routinely appear nowadays in a wide range of applications, due to modern data collection capabilities. Often such observations are taken over time, forming tensor time series. In this article we present a factor model approach to the analysis of high-dimensional dynamic tensor time series and multi-category dynamic transport networks. This article presents two estimation procedures along with their theoretical properties and simulation results. We present two applications to illustrate the model and its interpretations.

AB - Large tensor (multi-dimensional array) data routinely appear nowadays in a wide range of applications, due to modern data collection capabilities. Often such observations are taken over time, forming tensor time series. In this article we present a factor model approach to the analysis of high-dimensional dynamic tensor time series and multi-category dynamic transport networks. This article presents two estimation procedures along with their theoretical properties and simulation results. We present two applications to illustrate the model and its interpretations.

KW - Autocovariance matrices

KW - Cross-covariance matrices

KW - Dimension reduction

KW - Dynamic transport network

KW - Eigen-analysis

KW - Factor models

KW - Import–export

KW - Tensor time series

KW - Traffic

KW - Unfolding

UR - http://www.scopus.com/inward/record.url?scp=85106344670&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85106344670&partnerID=8YFLogxK

U2 - https://doi.org/10.1080/01621459.2021.1912757

DO - https://doi.org/10.1080/01621459.2021.1912757

M3 - Article

SN - 0162-1459

VL - 117

SP - 94

EP - 116

JO - Journal of the American Statistical Association

JF - Journal of the American Statistical Association

IS - 537

ER -

Factor Models for High-Dimensional Tensor Time Series

Abstract

ASJC Scopus subject areas

Keywords

Access to Document

Other files and links

Fingerprint

Cite this