Hypothesis Testing for the Covariance Matrix in High-Dimensional Transposable Data with Kronecker Product Dependence Structure
- Submitting institution
-
University of Brighton
- Unit of assessment
- 11 - Computer Science and Informatics
- Output identifier
- 7155403
- Type
- D - Journal article
- DOI
-
10.5705/ss.202018.0268
- Title of journal
- Statistica Sinica
- Article number
- -
- First page
- 1
- Volume
- 31
- Issue
- 3
- ISSN
- 1017-0405
- Open access status
- Exception within 3 months of publication
- Month of publication
- October
- Year of publication
- 2019
- URL
-
-
- Supplementary information
-
-
- Request cross-referral to
- -
- Output has been delayed by COVID-19
- No
- COVID-19 affected output statement
- -
- Forensic science
- No
- Criminology
- No
- Interdisciplinary
- No
- Number of additional authors
-
2
- Research group(s)
-
-
- Citation count
- -
- Proposed double-weighted
- No
- Reserve for an output with double weighting
- No
- Additional information
- Testing a large covariance matrix is essential when drawing inference in high-dimensional studies with non-normally distributed matrix-valued data because of the complex dependence structure implied by the row and column variables forming the data. This paper is significant because it proposes computationally inexpensive and powerful tests against a spherical, identity, diagonal, or fixed known covariance matrix without ignoring the complex dependence structure. Applications to genetics and electroencephalogram datasets indicate that analyses in Yin and Li (JMVA 2012), Xia and Li (BIOM 2016) and Ning and Liu (BIOMET 2013) gave misleading results.
- Author contribution statement
- -
- Non-English
- No
- English abstract
- -