Envelopes and partial least squares regression

• 成果类型: Article
• 署名作者: Cook, R. D.; Helland, I. S.; Su, Z.
• 署名单位: University of Minnesota System; University of Minnesota Twin Cities; University of Oslo; State University System of Florida; University of Florida
• 刊物名称: JOURNAL OF THE ROYAL STATISTICAL SOCIETY SERIES B-STATISTICAL METHODOLOGY
• ISSN/ISSBN: 1369-7412
• DOI: 10.1111/rssb.12018
• 发表日期: 2013
• 页码: 851-877
• 关键词: dimension reduction; prediction

摘要

We build connections between envelopes, a recently proposed context for efficient estimation in multivariate statistics, and multivariate partial least squares (PLS) regression. In particular, we establish an envelope as the nucleus of both univariate and multivariate PLS, which opens the door to pursuing the same goals as PLS but using different envelope estimators. It is argued that a likelihood-based envelope estimator is less sensitive to the number of PLS components that are selected and that it outperforms PLS in prediction and estimation.