作者: Ph. Besse , A. de Falguerolles
DOI: 10.1007/978-3-642-52468-4_11
关键词:
摘要: This paper investigates the problem of choice dimension in Principal Component Analysis (PCA). PCA is introduced as a model; loss function assessing stability fit considered. The then amounts to minimisation an expected which has be estimated. achieved by resampling methods. Different bootstrap and jackknife estimates are presented. behaviour these investigated on artificial data real data. resulting choices confronted with those given naive rules.