Sparse Principal Component Analysis in Hilbert Space

Xin Qi and Ruiyan Luo

Scandinavian Journal of Statistics, 2015, vol. 42, issue 1, 270-289

Abstract: type="main" xml:id="sjos12106-abs-0001"> Technical advances in many areas have produced more complicated high-dimensional data sets than the usual high-dimensional data matrix, such as the fMRI data collected in a period for independent trials, or expression levels of genes measured in different tissues. Multiple measurements exist for each variable in each sample unit of these data. Regarding the multiple measurements as an element in a Hilbert space, we propose Principal Component Analysis (PCA) in Hilbert space. The principal components (PCs) thus defined carry information about not only the patterns of variations in individual variables but also the relationships between variables. To extract the features with greatest contributions to the explained variations in PCs for high-dimensional data, we also propose sparse PCA in Hilbert space by imposing a generalized elastic-net constraint. Efficient algorithms to solve the optimization problems in our methods are provided. We also propose a criterion for selecting the tuning parameter.

Date: 2015
References: View references in EconPapers View complete reference list from CitEc
Citations: View citations in EconPapers (3)

Downloads: (external link)
http://hdl.handle.net/10.1111/sjos.12106 (text/html)
Access to full text is restricted to subscribers.

Related works:
This item may be available elsewhere in EconPapers: Search for items with the same title.

Export reference: BibTeX RIS (EndNote, ProCite, RefMan) HTML/Text

Persistent link: https://EconPapers.repec.org/RePEc:bla:scjsta:v:42:y:2015:i:1:p:270-289

Ordering information: This journal article can be ordered from
http://www.blackwell ... bs.asp?ref=0303-6898

Access Statistics for this article

Scandinavian Journal of Statistics is currently edited by ÿrnulf Borgan and Bo Lindqvist

More articles in Scandinavian Journal of Statistics from Danish Society for Theoretical Statistics, Finnish Statistical Society, Norwegian Statistical Association, Swedish Statistical Association
Bibliographic data for series maintained by Wiley Content Delivery ().