蔡天文 Tony Cai 2014-10-10 6:03 PM

Sparse PCA: Optimal Rates and Adaptive Estimation

Abstract

Principal component analysis (PCA) is one of the most commonly used statistical procedures with a wide range of applications.This paper considers both minimax and adaptive estimation of the principal subspace in the high dimensional setting. Under mild technical conditions, we first establish the optimal rates of convergence for estimating the principal subspace which are sharp with respect to all the parameters, thus providing a complete characterization of the difficulty of the estimation problem in term of the convergence rate. The lower bound is obtained by calculating the local metric entropy and an application of Fano's Lemma. The rate optimal estimator is constructed using aggregation, which, however, might not be computationally feasible.

 

We then introduce an adaptive procedure for estimating the principal subspace which is fully data driven and can be computed efficiently. It is shown that the estimator attains the optimal rates of convergence simultaneously over a large collection of the parameter spaces. A key idea in our construction is a reduction scheme which reduces the sparse PCA problem to a high-dimensional multivariate regression problem. This method is potentially also useful for other related problems.


Full Article


KEYWORDS

SHARE & LIKE

COMMENTS

ABOUT THE AUTHOR

蔡天文 Tony Cai

宾夕法尼亚大学沃顿商学院Dorothy Silberberg 统计学讲席教授、应用数学及计算科学教授

0 Following 12 Fans 0 Projects 7 Articles

SIMILAR ARTICLES

AbstractPrincipal component analysis (PCA) is one of the most commonly used statistical procedures with a wide range of applications.This paper conside

Read More

AbstractEstimation of low-rank matrices is of significant interest in a range of contemporary applications. In this paper, we introduce a rank-one proj

Read More

AbstractIt is often of interest to understand how the structure of a genetic network differs between two conditions. In this paper, each condition-spec

Read More

AbstractThis paper considers testing the equality of multiple high-dimensional mean vectors under dependency. We propose a test that is based on a line

Read More

AbstractPrincipal component analysis (PCA) is one of the most commonly used statistical procedures for dimension reduction. This paper presents some re

Read More

AbstractThis paper studies the asymptotic behaviors of the pairwise angles among n randomly and uniformly distributed unit vectors in ℜp as the number

Read More

AbstractTolerance intervals are widely used in industrial applications. So far attention has been mainly focused on the construction of tolerance inter

Read More