Estimating Common Principal Components in High Dimensions

Abstract

We consider the problem of minimizing an objective function that depends on an orthonormal matrix. This situation is encountered when looking for common principal components, for example, and the Flury method is a popular approach. However, the Flury method is not effective for higher dimensional problems. We obtain several simple majorization-minizmation (MM) algorithms that provide solutions to this problem and are effective in higher dimensions. We then use simulated data to compare them with other approaches in terms of convergence and computational time.

Cited by in corpus (14)

• A Mixture of Generalized Hyperbolic Distributions
• Parsimonious Skew Mixture Models for Model-Based Clustering and Classification
• A Mixture of Generalized Hyperbolic Factor Analyzers
• Unsupervised Learning via Mixtures of Skewed Distributions with Hypercube Contours
• Model Based Clustering of High-Dimensional Binary Data
• Mixture Model Averaging for Clustering
• Parsimonious Shifted Asymmetric Laplace Mixtures
• A robust approach to model-based classification based on trimming and constraints
• Flexible Clustering for High-Dimensional Data via Mixtures of Joint Generalized Hyperbolic Models
• Clustering Higher Order Data: An Application to Pediatric Multi-variable Longitudinal Data
• Hypothesis Testing for Parsimonious Gaussian Mixture Models
• Multivariate normal mixture modeling, clustering and classification with the rebmix package
• Improving Model Choice in Classification: An Approach Based on Clustering of Covariance Matrices
• A parsimonious family of multivariate Poisson-lognormal distributions for clustering multivariate count data