Efficient estimation of approximate factor models via penalized maximum likelihood

DOI10.1016/J.JECONOM.2015.10.003zbMATH Open1390.62107arXiv1209.5911OpenAlexW3124904104MaRDI QIDQ898581FDOQ898581

Authors: Jushan Bai, Yuan Liao 

Publication date: 18 December 2015

Published in: Journal of Econometrics (Search for Journal in Brave)

Abstract: We study the estimation of a high dimensional approximate factor model in the presence of both cross sectional dependence and heteroskedasticity. The classical method of principal components analysis (PCA) does not efficiently estimate the factor loadings or common factors because it essentially treats the idiosyncratic error to be homoskedastic and cross sectionally uncorrelated. For efficient estimation it is essential to estimate a large error covariance matrix. We assume the model to be conditionally sparse, and propose two approaches to estimating the common factors and factor loadings; both are based on maximizing a Gaussian quasi-likelihood and involve regularizing a large covariance sparse matrix. In the first approach the factor loadings and the error covariance are estimated separately while in the second approach they are estimated jointly. Extensive asymptotic analysis has been carried out. In particular, we develop the inferential theory for the two-step estimation. Because the proposed approaches take into account the large error covariance matrix, they produce more efficient estimators than the classical PCA methods or methods based on a strict factor model.

Full work available at URL: https://arxiv.org/abs/1209.5911

Recommendations

• On the penalized maximum likelihood estimation of high-dimensional approximate factor model
• Efficient estimation of factor models
• Statistical analysis of factor models of high dimension
• Statistical analysis of sparse approximate factor models
• Quasi maximum likelihood analysis of high dimensional constrained factor models

zbMATH Keywords

principal componentsSCADheteroskedasticitypenalized maximum likelihoodhigh dimensionalitythresholdingadaptive LASSOconditional sparsecross-sectional correlationunknown factors

Mathematics Subject Classification ID

Factor analysis and principal components; correspondence analysis (62H25) Ridge regression; shrinkage estimators (Lasso) (62J07) Economic time series analysis (91B84)

Cites Work

• Sparse estimation of a covariance matrix
• A penalized matrix decomposition, with applications to sparse principal components and canonical correlation analysis
• The Adaptive Lasso and Its Oracle Properties
• Variable Selection via Nonconcave Penalized Likelihood and its Oracle Properties
• Forecasting Using Principal Components From a Large Number of Predictors
• Title not available (Why is that?)
• Title not available (Why is that?)
• Inferential Theory for Factor Models of Large Dimensions
• Determining the Number of Factors in Approximate Factor Models
• The Generalized Dynamic Factor Model
• A two-step estimator for large approximate dynamic factor models based on Kalman filtering
• High dimensional covariance matrix estimation using a factor model
• Factor modeling for high-dimensional time series: inference for the number of factors
• Determining the Number of Factors in the General Dynamic Factor Model
• Large Covariance Estimation by Thresholding Principal Orthogonal Complements
• Consistent Estimates Based on Partially Consistent Observations
• Improved penalization for determining the number of factors in approximate factor models
• Covariance regularization by thresholding
• Sparsistency and rates of convergence in large covariance matrix estimation
• Nonlinear shrinkage estimation of large-dimensional covariance matrices
• Sparse inverse covariance estimation with the graphical lasso
• Sparse permutation invariant covariance estimation
• Adaptive Lasso for sparse high-dimensional regression models
• Asymptotics of the principal components estimator of large factor models with weakly influential factors
• Statistical analysis of factor models of high dimension
• Constrained Factor Models
• A Testing Procedure for Determining the Number of Factors in Approximate Factor Models With Large Datasets
• PCA consistency in high dimension, low sample size context
• High-dimensional covariance estimation by minimizing \(\ell _{1}\)-penalized log-determinant divergence
• Posterior contraction in sparse Bayesian factor models for massive covariance matrices
• High dimensional inverse covariance matrix estimation via linear programming
• Estimation of high-dimensional low-rank matrices
• Spectrum estimation for large dimensional covariance matrices using random matrix theory
• The adaptive and the thresholded Lasso for potentially misspecified models (and a lower bound for the Lasso)
• Positive definite estimators of large covariance matrices
• High-dimensional covariance estimation based on Gaussian graphical models
• Title not available (Why is that?)
• Positive-definite \(\ell_1\)-penalized estimation of large covariance matrices
• The DC (Difference of convex functions) programming and DCA revisited with DC models of real world nonconvex optimization problems
• High-dimensional covariance matrix estimation in approximate factor models
• GLS estimation of dynamic factor models
• Efficient estimation of factor models
• Determining the number of global and country-specific factors in the euro area

Cited In (25)

• Factor Extraction in Dynamic Factor Models: Kalman Filter Versus Principal Components
• Single- and multiple-group penalized factor analysis: a trust-region algorithm approach with integrated automatic multiple tuning parameter selection
• Shrinkage estimation of dynamic panel data models with interactive fixed effects
• Efficient estimation of factor models
• Prediction in functional regression with discretely observed and noisy covariates
• Preprocessing noisy functional data: a multivariate perspective
• Sparse estimation via nonconcave penalized likelihood in factor analysis model
• Mining the factor zoo: estimation of latent factor models with sufficient proxies
• Estimating a covariance matrix for market risk management and the case of credit default swaps
• Semiparametric model for covariance regression analysis
• Panel threshold models with interactive fixed effects
• Penalized Regression for Multiple Types of Many Features With Missing Data
• Robust inference of risks of large portfolios
• Shrinkage Estimation of Factor Models With Global and Group-Specific Factors
• Interpretable Sparse Proximate Factors for Large Dimensions
• Multi-population mortality modeling: when the data is too much and not enough
• On the statistical analysis of high-dimensional factor models
• A spatial panel quantile model with unobserved heterogeneity
• Improved penalization for determining the number of factors in approximate factor models
• Testing for time-varying factor loadings in high-dimensional factor models
• Statistical analysis of sparse approximate factor models
• Consistently recovering the signal from noisy functional data
• Least squares estimation of large dimensional threshold factor models
• Approximate factor models with weaker loadings
• On the penalized maximum likelihood estimation of high-dimensional approximate factor model

Uses Software

• glasso
• spcov
• PMA

This page was built for publication: Efficient estimation of approximate factor models via penalized maximum likelihood

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q898581)