An alternating minimization algorithm for factor analysis.

DOI10.14736/KYB-2019-4-0740MaRDI QIDQ5218998zbMATH OpenOpenAlexFDO

Authors Valentina Ciccone, Augusto Ferrante, Mattia Zorzi

Publication date 6 March 2020

Published in Kybernetika (Search for Journal in Brave)

Full work available at URL https://arxiv.org/abs/1806.04433

zbMATH Keywords

covariance matricesfactor analysislow rank matricesmatrix decompositionprojections

Mathematics Subject Classification ID

Estimation in multivariate analysis (62H12) Factor analysis and principal components; correspondence analysis (62H25) Analysis of variance and covariance (ANOVA) (62J10)

Abstract: The problem of decomposing a given covariance matrix as the sum of a positive semi-definite matrix of given rank and a positive semi-definite diagonal matrix, is considered. We present a projection-type algorithm to address this problem. This algorithm appears to perform extremely well and is extremely fast even when the given covariance matrix has a very large dimension. The effectiveness of the algorithm is assessed through simulation studies and by applications to three real datasets that are considered as benchmark for the problem. A local convergence analysis of the algorithm is also presented.

Recommendations

• Factor Analysis and Alternating Minimization
• Factor analysis models via I-divergence optimization
• Certifiably optimal low rank factor analysis
• On a DC-optimization-problem from statistical factor analysis
• On the Minres Method of Factor Analysis

Cites work

• A Structure Theory for Linear Dynamic Errors-in-Variables Models
• Alternating Projections on Manifolds
• AR Identification of Latent-Variable Graphical Models
• Certifiably optimal low rank factor analysis
• Critical points of matrix least squares distance functions
• Determining the Number of Factors in Approximate Factor Models
• Factor Models With Real Data: A Robust Estimation of the Number of Factors
• scientific article; zbMATH DE number 3605818 (Why is no real title available?)
• Identifiability of a Linear Relation between Variables Which Are Subject to Error
• Linear algebra and its applications
• Linear models based on noisy data and the Frisch scheme
• Minimum rank and minimum trace of covariance matrices
• Modelling high-dimensional time series by generalized linear dynamic factor models: an introductory survey
• Noisy matrix decomposition via convex relaxation: optimal rates in high dimensions
• Rank-reducibility of a symmetric matrix and sampling theory of minimum trace factor analysis
• Some necessary conditions for common-factor analysis
• Statistical analysis of factor models of high dimension
• Statistical inference of minimum rank factor analysis
• System Identification by Dynamic Factor Models

Cited in

(14)

• Computation of the maximum likelihood estimator in low-rank factor analysis
• Approximate low-rank factorization with structured factors
• An alternating least squares algorithm for PARAFAC2 and three-way DEDICOM
• Construction of covariance matrices with a specified discrepancy function minimizer, with application to factor analysis
• Factor analysis models via I-divergence optimization
• A Unified Algorithm for Multivariate Analytic Factorization
• On the Minres Method of Factor Analysis
• A Projection Decomposition for Bivariate Discrete Probability Distributions
• scientific article; zbMATH DE number 4064282 (Why is no real title available?)
• Certifiably optimal low rank factor analysis
• Factor Analysis and Alternating Minimization
• A very fast algorithm for matrix factorization
• Low-rank plus diagonal approximations for Riccati-like matrix differential equations
• On the Wiberg algorithm for matrix factorization in the presence of missing components

This page was built for publication: An alternating minimization algorithm for factor analysis.

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q5218998)