Order Determination for Spiked Models

TL;DR

This paper introduces a new, easy-to-implement criterion for determining the number of spikes in large-dimensional matrices, crucial for dimension reduction and signal detection, with demonstrated effectiveness across various models.

Contribution

It proposes a novel 'valley-cliff' criterion for order determination in high-dimensional spiked models, addressing challenges posed by asymptotic eigenvalue behaviors.

Findings

The method accurately identifies the number of spikes in simulated data.

It outperforms existing methods in finite sample scenarios.

The approach is versatile across different matrix models.

Abstract

Motivated by dimension reduction in regression analysis and signal detection, we investigate the order determination for large dimension matrices including spiked models of which the numbers of covariates are proportional to the sample sizes for different models. Because the asymptotic behaviour of the estimated eigenvalues of the corresponding matrices differ completely from those in fixed dimension scenarios, we then discuss the largest possible number we can identify and introduce a "valley-cliff" criterion. We propose two versions of the criterion: one based on the original differences of eigenvalues and the other based on the transformed differences, which reduces the effects of ridge selection in the former one. This generic method is very easy to implement and computationally inexpensive, and it can be applied to various matrices. As examples, we focus on spiked population models, spiked Fisher matrices and factor models with auto-covariance matrices. Numerical studies are conducted to examine the method's finite sample performances and to compare it with existing methods.