Partial least squares for dependent data

• 成果类型: Article
• 署名作者: Singer, Marco; Krivobokova, Tatyana; Munk, Axel; De Groot, Bert
• 署名单位: University of Gottingen; Max Planck Society
• 刊物名称: BIOMETRIKA
• ISSN/ISSBN: 0006-3444
• DOI: 10.1093/biomet/asw010
• 发表日期: 2016
• 页码: 351362
• 关键词: regression; proteins; PLS; chemometrics; prediction; matrices; MOTIONS; models

摘要

We consider the partial least squares algorithm for dependent data and study the consequences of ignoring the dependence both theoretically and numerically. Ignoring nonstationary dependence structures can lead to inconsistent estimation, but a simple modification yields consistent estimation. A protein dynamics example illustrates the superior predictive power of the proposed method.