Certifiably optimal low rank factor analysis
From MaRDI portal
Abstract: Factor Analysis (FA) is a technique of fundamental importance that is widely used in classical and modern multivariate statistics, psychometrics and econometrics. In this paper, we revisit the classical rank-constrained FA problem, which seeks to approximate an observed covariance matrix (), by the sum of a Positive Semidefinite (PSD) low-rank component () and a diagonal matrix () (with nonnegative entries) subject to being PSD. We propose a flexible family of rank-constrained, nonlinear Semidefinite Optimization based formulations for this task. We introduce a reformulation of the problem as a smooth optimization problem with convex compact constraints and propose a unified algorithmic framework, utilizing state of the art techniques in nonlinear optimization to obtain high-quality feasible solutions for our proposed formulation. At the same time, by using a variety of techniques from discrete and global optimization, we show that these solutions are certifiably optimal in many cases, even for problems with thousands of variables. Our techniques are general and make no assumption on the underlying problem data. The estimator proposed herein, aids statistical interpretability, provides computational scalability and significantly improved accuracy when compared to current, publicly available popular methods for rank-constrained FA. We demonstrate the effectiveness of our proposal on an array of synthetic and real-life datasets. To our knowledge, this is the first paper that demonstrates how a previously intractable rank-constrained optimization problem can be solved to provable optimality by coupling developments in convex analysis and in discrete optimization.
Recommendations
- Computation of the maximum likelihood estimator in low-rank factor analysis
- An alternating minimization algorithm for factor analysis.
- On a DC-optimization-problem from statistical factor analysis
- Statistical inference of minimum rank factor analysis
- The analysis of multivariate data using semi-definite programming
Cited in
(18)- Flexible low-rank statistical modeling with missing data and side information
- Factor analysis procedures revisited from the comprehensive model with unique factors decomposed into specific factors and errors
- A look at robustness and stability of \(\ell_1\)-versus \(\ell_0\)-regularization: discussion of papers by Bertsimas et al. and Hastie et al.
- Certifiably optimal sparse principal component analysis
- Rank regularized estimation of approximate factor models
- Computation of the maximum likelihood estimator in low-rank factor analysis
- Factor analysis models via I-divergence optimization
- Mixed-Projection Conic Optimization: A New Paradigm for Modeling Rank Constraints
- An alternating minimization algorithm for factor analysis.
- A new perspective on low-rank optimization
- Exterior-point optimization for sparse and low-rank optimization
- Optimal ensemble construction for multistudy prediction with applications to mortality estimation
- Low-rank plus diagonal approximations for Riccati-like matrix differential equations
- Interpreting latent variables in factor models via convex optimization
- Latent Gaussian dynamic factor modeling and forecasting for multivariate count time series
- Hierarchical relations among principal component and factor analysis procedures elucidated from a comprehensive model
- Computational complexity of decomposing a symmetric matrix as a sum of positive semidefinite and diagonal matrices
- Improved approximation algorithms for orthogonally constrained problems using semidefinite optimization
This page was built for publication: Certifiably optimal low rank factor analysis
Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q5361290)
