2022
DOI: 10.1016/j.jeconom.2021.04.001
|View full text |Cite
|
Sign up to set email alerts
|

Projected estimation for large-dimensional matrix factor models

Help me understand this report

Search citation statements

Order By: Relevance

Paper Sections

Select...
3

Citation Types

7
76
0

Year Published

2022
2022
2024
2024

Publication Types

Select...
4
1
1

Relationship

1
5

Authors

Journals

citations
Cited by 36 publications
(83 citation statements)
references
References 22 publications
7
76
0
Order By: Relevance
“…Wang et al (2019) propose estimators of the factor loading matrices and of numbers of the row and column factors based on the eigen-analysis of the auto-cross-covariance matrix, under the assumption that the idiosyncratic term E t is white noise. Chen and Fan (2021) propose an α-PCA method for inference of (1.1), which conducts eigen-analysis of a weighted average of the sample mean and the column (row) sample covariance matrix; Yu et al (2021) proposed a projected estimation method which further improved the estimation efficiency of the factor loading matrices and the numbers of factors. He et al (2021) proposed a strong rule to determine whether there is a factor structure of matrix time series and also propose a sequential procedure to determine the numbers of factors.…”
Section: Introductionmentioning
confidence: 99%
See 4 more Smart Citations
“…Wang et al (2019) propose estimators of the factor loading matrices and of numbers of the row and column factors based on the eigen-analysis of the auto-cross-covariance matrix, under the assumption that the idiosyncratic term E t is white noise. Chen and Fan (2021) propose an α-PCA method for inference of (1.1), which conducts eigen-analysis of a weighted average of the sample mean and the column (row) sample covariance matrix; Yu et al (2021) proposed a projected estimation method which further improved the estimation efficiency of the factor loading matrices and the numbers of factors. He et al (2021) proposed a strong rule to determine whether there is a factor structure of matrix time series and also propose a sequential procedure to determine the numbers of factors.…”
Section: Introductionmentioning
confidence: 99%
“…In the current work, we first propose least squares estimators for the matrix factor models. The most interesting finding is that the least squares estimators are equivalent to the projected estimators by Yu et al (2021), i.e., the projected estimators minimizes the least squares loss function. This finding provides another rationale for the projected estimation procedure by Yu et al (2021), which is initially proposed for reducing the magnitudes of the idiosyncratic error components and thereby increasing the signal-to-noise ratio.…”
Section: Introductionmentioning
confidence: 99%
See 3 more Smart Citations