manuscript-R.pdf (4 MB)

Principal weighted support vector machines for sufficient dimension reduction in binary classification

Download (4 MB)
journal contribution
posted on 23.03.2021, 14:23 by SJ Shin, Y Wu, HH Zhang, Y Liu
© 2017 Biometrika Trust. Sufficient dimension reduction is popular for reducing data dimensionality without stringent model assumptions. However, most existing methods may work poorly for binary classification. For example, sliced inverse regression (Li, 1991) can estimate at most one direction if the response is binary. In this paper we propose principal weighted support vector machines, a unified framework for linear and nonlinear sufficient dimension reduction in binary classification. Its asymptotic properties are studied, and an efficient computing algorithm is proposed. Numerical examples demonstrate its performance in binary classification.

History

Publisher Statement

This is a pre-copyedited, author-produced version of an article accepted for publication in Biometrika following peer review. The version of record: Shin, S. J., Wu, Y., Zhang, H. H.Liu, Y. (2017). Principal weighted support vector machines for sufficient dimension reduction in binary classification. Biometrika, 104(1), 67-81. is available online at: https://doi.org/10.1093/biomet/asw057.

Citation

Shin, S. J., Wu, Y., Zhang, H. H.Liu, Y. (2017). Principal weighted support vector machines for sufficient dimension reduction in binary classification. Biometrika, 104(1), 67-81. https://doi.org/10.1093/biomet/asw057

Publisher

Oxford University Press (OUP)

Language

en

issn

0006-3444

Exports