Statistical analysis of factor models of high dimension
arXiv:1205.6617 · doi:10.1214/11-AOS966
Abstract
This paper considers the maximum likelihood estimation of factor models of high dimension, where the number of variables (N) is comparable with or even greater than the number of observations (T). An inferential theory is developed. We establish not only consistency but also the rate of convergence and the limiting distributions. Five different sets of identification conditions are considered. We show that the distributions of the MLE estimators depend on the identification restrictions. Unlike the principal components approach, the maximum likelihood estimator explicitly allows heteroskedasticities, which are jointly estimated with other parameters. Efficiency of MLE relative to the principal components method is also considered.
arXiv commentPublished in at http://dx.doi.org/10.1214/11-AOS966 the Annals of Statistics (http://www.imstat.org/aos/) by the Institute of Mathematical Statistics (http://www.imstat.org)
Cited by in corpus (44)
- Projected principal component analysis in factor models
- Theory and methods of panel data models with interactive effects
- Deterministic parallel analysis: An improved method for selecting factors and principal components
- Identifying the number of factors from singular values of a large sample auto-covariance matrix
- Infinite Mixtures of Infinite Factor Analysers
- Large Covariance Estimation through Elliptical Factor Models
- Confounder Adjustment in Multiple Hypothesis Testing
- Recent Developments on Factor Models and its Applications in Econometric Learning
- Expandable Factor Analysis
- Principal component analysis for big data
- Item Response Theory -- A Statistical Framework for Educational and Psychological Measurement
- Next Generation Models for Portfolio Risk Management: An Approach Using Financial Big Data
- Semi-parametric TEnsor Factor Analysis by Iteratively Projected Singular Value Decomposition
- Sufficient and Necessary Conditions for the Identifiability of the -matrix
- Low-rank covariance matrix estimation for factor analysis in anisotropic noise: application to array processing and portfolio selection
- Multi-Scale Factor Analysis of High-Dimensional Brain Signals
- Optimal index insurance and basis risk decomposition: an application to Kenya
- Certifiably Optimal Low Rank Factor Analysis
- Towards Clarifying the Theory of the Deconfounder
- Factor Models for High-Dimensional Tensor Time Series
- Robust high dimensional factor models with applications to statistical machine learning
- Statistical Inferences Using Large Estimated Covariances for Panel Data and Factor Models
- Learning Latent Factors from Diversified Projections and its Applications to Over-Estimated and Weak Factors
- Identifiability of Hierarchical Latent Attribute Models
- User-Friendly Covariance Estimation for Heavy-Tailed Distributions
- Modeling High-Dimensional Unit-Root Time Series
- A two-way factor model for high-dimensional matrix data
- Factor analysis in high dimensional biological data with dependent observations
- Consistently recovering the signal from noisy functional data
- Large-dimensional Factor Analysis without Moment Constraints
- FarmTest: Factor-Adjusted Robust Multiple Testing with Approximate False Discovery Control
- Adaptive Estimation in Structured Factor Models with Applications to Overlapping Clustering
- A Note on the Likelihood Ratio Test in High-Dimensional Exploratory Factor Analysis
- Sparse Approximate Factor Estimation for High-Dimensional Covariance Matrices
- Estimating and accounting for unobserved covariates in high dimensional correlated data
- Approximate factor analysis model building via alternating I-divergence minimization
- Robust Principal Component Analysis with Non-Sparse Errors
- Estimation for ultra-high dimensional factor model: a pivotal variable detection based approach
- Detecting approximate replicate components of a high-dimensional random vector with latent structure
- Augmented Factor Models with Applications to Validating Market Risk Factors and Forecasting Bond Risk Premia
- SOFAR: large-scale association network learning
- Network-Assisted Estimation for Large-dimensional Factor Model with Guaranteed Convergence Rate Improvement
- Bi-cross-validation for factor analysis
- Heterotic Risk Models