Computation of the maximum likelihood estimator in low-rank factor analysis
From MaRDI portal
Publication:2425172
Abstract: Factor analysis, a classical multivariate statistical technique is popularly used as a fundamental tool for dimensionality reduction in statistics, econometrics and data science. Estimation is often carried out via the Maximum Likelihood (ML) principle, which seeks to maximize the likelihood under the assumption that the positive definite covariance matrix can be decomposed as the sum of a low rank positive semidefinite matrix and a diagonal matrix with nonnegative entries. This leads to a challenging rank constrained nonconvex optimization problem. We reformulate the low rank ML Factor Analysis problem as a nonlinear nonsmooth semidefinite optimization problem, study various structural properties of this reformulation and propose fast and scalable algorithms based on difference of convex (DC) optimization. Our approach has computational guarantees, gracefully scales to large problems, is applicable to situations where the sample covariance matrix is rank deficient and adapts to variants of the ML problem with additional constraints on the problem parameters. Our numerical experiments demonstrate the significant usefulness of our approach over existing state-of-the-art approaches.
Recommendations
- Certifiably optimal low rank factor analysis
- Maximum likelihood factor analysis with rank-deficient sample covariance matrices
- An alternating minimization algorithm for factor analysis.
- On maximum likelihood estimation in factor analysis-an algebraic derivation
- Maximum likelihood for matrices with rank constraints
Cites work
- scientific article; zbMATH DE number 3950216 (Why is no real title available?)
- scientific article; zbMATH DE number 3673370 (Why is no real title available?)
- scientific article; zbMATH DE number 1964693 (Why is no real title available?)
- scientific article; zbMATH DE number 2107836 (Why is no real title available?)
- scientific article; zbMATH DE number 757676 (Why is no real title available?)
- scientific article; zbMATH DE number 6159604 (Why is no real title available?)
- scientific article; zbMATH DE number 3258664 (Why is no real title available?)
- scientific article; zbMATH DE number 3396952 (Why is no real title available?)
- All convex invariant functions of hermitian matrices
- Certifiably optimal low rank factor analysis
- Computing B-stationary points of nonsmooth DC programs
- Conic optimization via operator splitting and homogeneous self-dual embedding
- Convex Analysis on the Hermitian Matrices
- Convex analysis and nonlinear optimization. Theory and examples.
- Derivatives of Spectral Functions
- Diagonal and low-rank matrix decompositions, correlation matrices, and ellipsoid fitting
- Difference-of-convex learning: directional stationarity, optimality, and sparsity
- EM algorithms for ML factor analysis
- Fast low-rank modifications of the thin singular value decomposition
- Latent variable models and factor analysis. A unified approach
- Maximum likelihood factor analysis with rank-deficient sample covariance matrices
- Model selection through sparse maximum likelihood estimation for multivariate Gaussian or binary data
- On the pervasiveness of difference-convexity in optimization and statistics
- Pattern recognition and machine learning.
- Some contributions to maximum likelihood factor analysis
- Sparse inverse covariance estimation with the graphical lasso
- Statistical analysis of factor models of high dimension
- Statistical inference of minimum rank factor analysis
- The Concave-Convex Procedure
- Variational Analysis
Cited in
(5)- scientific article; zbMATH DE number 7415123 (Why is no real title available?)
- A Matrix-Free Likelihood Method for Exploratory Factor Analysis of High-Dimensional Gaussian Data
- Certifiably optimal low rank factor analysis
- Interpreting latent variables in factor models via convex optimization
- Maximum likelihood factor analysis with rank-deficient sample covariance matrices
This page was built for publication: Computation of the maximum likelihood estimator in low-rank factor analysis
Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q2425172)
