Factor models for matrix-valued high-dimensional time series

By Dong Wang, Xialu Liu, Rong Chen. Journal of Econometrics, 2016

188 citations recorded by the source corpus.

In finance, economics and many other fields, observations in a matrix form are often observed over time. For example, many economic indicators are obtained in different countries over time. Various financial characteristics of many companies are reported over time. Although it is natural to turn a matrix observation into a long vector then use standard vector time series models or factor analysis, it is often the case that the columns and rows of a matrix represent different sets of information that are closely interrelated in a very structural way. We propose a novel factor model that maintains and utilizes the matrix structure to achieve greater dimensional reduction as well as finding clearer and more interpretable factor structures. Estimation procedure and its theoretical properties are investigated and demonstrated with simulated and real examples.

Subfields: Artificial Intelligence, Economics and Econometrics, Finance

View Paper: https://www.semanticscholar.org/paper/9b6337bd13696ebe2caa5833af8e0d759688d32b

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Artificial Intelligence

1. Jinyuan Chang
2. Rong Chen
3. Lin Yang
4. Jingjing He
5. Han Xiao
6. Dan Yang
7. Qiwei Yao
8. Jianqing Fan

In Economics and Econometrics

1. Jinyuan Chang
2. Qiwei Yao
3. Rong Chen
4. Lin Yang
5. Jingjing He
6. Han Xiao
7. Clifford Lam
8. S. Ling

In Finance

1. Jinyuan Chang
2. Rong Chen
3. Qiwei Yao
4. Han Xiao
5. Dan Yang
6. S. Ling
7. Clifford Lam
8. Christian Francq

Citations in the corpus

• Autoregressive models for matrix-valued time series (2018, result): “This second approach is also related to the recent work (Wang et al., 2019), which studied the factor models for matrix-valued time series.”
• Modelling matrix time series via a tensor CP-decomposition (2021, methodology): “Most existing literature on matrix time series is based on the factor modelling via the Tucker decomposition; see Chen and Chen (2019), Wang et al. (2019) and Chen et al. (2020).”