Learning a factor model via regularized PCA

DOI10.1007/S10994-013-5345-8zbMATH Open1293.68230arXiv1111.6201OpenAlexW3098601467MaRDI QIDQ399883FDOQ399883

Authors: Yi-Hao Kao, Benjamin Van Roy 

Publication date: 20 August 2014

Published in: Machine Learning (Search for Journal in Brave)

Abstract: We consider the problem of learning a linear factor model. We propose a regularized form of principal component analysis (PCA) and demonstrate through experiments with synthetic and real data the superiority of resulting estimates to those produced by pre-existing factor analysis approaches. We also establish theoretical results that explain how our algorithm corrects the biases induced by conventional approaches. An important feature of our algorithm is that its computational requirements are similar to those of PCA, which enjoys wide use in large part due to its efficiency.

Full work available at URL: https://arxiv.org/abs/1111.6201

Recommendations

• Probabilistic Principal Component Analysis
• Practical approaches to principal component analysis in the presence of missing values
• Regularised PCA to denoise and visualise data
• Structural, Syntactic, and Statistical Pattern Recognition
• On consistency and sparsity for principal components analysis in high dimensions

zbMATH Keywords

factor modelhigh-dimensional dataprincipal component analysisregularizationcovariance matrix estimation

Mathematics Subject Classification ID

Factor analysis and principal components; correspondence analysis (62H25) Learning and adaptive systems in artificial intelligence (68T05) Analysis of variance and covariance (ANOVA) (62J10)

Cites Work

• Distributed optimization and statistical learning via the alternating direction method of multipliers
• Sparse inverse covariance estimation with the graphical lasso
• On the distribution of the largest eigenvalue in principal components analysis
• Robust factor analysis.
• Eigenvalues of large sample covariance matrices of spiked population models
• Robust principal component analysis?
• Model selection through sparse maximum likelihood estimation for multivariate Gaussian or binary data
• Model selection and estimation in the Gaussian graphical model
• Asymptotics of sample eigenstructure for a large dimensional spiked covariance model
• A Direct Formulation for Sparse PCA Using Semidefinite Programming
• Latent variable graphical model selection via convex optimization
• High-dimensional covariance estimation by minimizing \(\ell _{1}\)-penalized log-determinant divergence
• Probabilistic Principal Component Analysis
• EM algorithms for ML factor analysis
• Adaptive estimation of a quadratic functional by model selection.
• Title not available (Why is that?)
• Factor analysis and AIC
• Covariance estimation: the GLM and regularization perspectives
• High-dimensional analysis of semidefinite relaxations for sparse principal components

Cited In (6)

• Rigid transformations for stabilized lower dimensional space to support subsurface uncertainty quantification and interpretation
• Directed principal component analysis
• Decision-based model selection
• Regularised PCA to denoise and visualise data
• Learning linear PCA with convex semi-definite programming
• Sparse factor analysis for learning and content analytics

Uses Software

• glasso

This page was built for publication: Learning a factor model via regularized PCA

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q399883)