Inference for low-rank models
From MaRDI portal
Abstract: This paper studies inference in linear models with a high-dimensional parameter matrix that can be well-approximated by a ``spiked low-rank matrix. A spiked low-rank matrix has rank that grows slowly compared to its dimensions and nonzero singular values that diverge to infinity. We show that this framework covers a broad class of models of latent-variables which can accommodate matrix completion problems, factor models, varying coefficient models, and heterogeneous treatment effects. For inference, we apply a procedure that relies on an initial nuclear-norm penalized estimation step followed by two ordinary least squares regressions. We consider the framework of estimating incoherent eigenvectors and use a rotation argument to argue that the eigenspace estimation is asymptotically unbiased. Using this framework we show that our procedure provides asymptotically normal inference and achieves the semiparametric efficiency bound. We illustrate our framework by providing low-level conditions for its application in a treatment effects context where treatment assignment might be strongly dependent.
Recommendations
Cites work
- A simpler approach to matrix completion
- De-Biased Sparse PCA: Inference for Eigenstructure of Large Covariance Matrices
- Efficient estimation of linear functionals of principal components
- Entrywise eigenvector analysis of random matrices with low expected rank
- Estimating Number of Factors by Adjusted Eigenvalues Thresholding
- Estimation of (near) low-rank matrices with noise and high-dimensional scaling
- Estimation of high-dimensional low-rank matrices
- Exact matrix completion via convex optimization
- Guaranteed Matrix Completion via Non-Convex Factorization
- High-Dimensional Principal Component Analysis with Heterogeneous Missingness
- scientific article; zbMATH DE number 3169866 (Why is no real title available?)
- Inference and uncertainty quantification for noisy matrix completion
- Inference for low-rank models
- Least squares after model selection in high-dimensional sparse models
- Low-rank matrix completion using alternating minimization
- Matrix completion and low-rank SVD via fast alternating least squares
- Matrix Completion From a Few Entries
- Matrix Completion Methods for Causal Panel Data Models
- Minimax bounds for sparse PCA with noisy high-dimensional data
- Noisy low-rank matrix completion with general sampling distribution
- Noisy matrix completion: understanding statistical guarantees for convex relaxation via nonconvex optimization
- Nuclear-norm penalization and optimal rates for noisy low-rank matrix completion
- Optimal detection of sparse principal components in high dimension
- Recovering Low-Rank Matrices From Few Coefficients in Any Basis
- Semiparametric efficiency bounds for high-dimensional models
- Sparse PCA: optimal rates and adaptive estimation
- Statistical Inferences of Linear Forms for Noisy Matrix Completion
Cited in
(19)- Universal rank inference via residual subsampling with application to large networks
- Inference for low-rank models
- Inference for low-rank completion without sample splitting with application to treatment effect estimation
- Inference for heteroskedastic PCA with missing data
- Statistical inference for matrix-vector linear regression without debiasing under Kronecker covariance structure
- Sieve estimation of state-varying factor models
- Inference for large dimensional factor models under general missing data patterns
- High-dimensional conditional factor model
- Statistical Quantile Learning for Large Additive Latent Variable Models
- Online Policy Learning and Inference by Matrix Completion
- Inference for Low-Rank Models Without Estimating the Rank
- Inference of dependency knowledge graph for electronic health records
- Estimation and inference for unbalanced panel data models with interactive fixed effects
- Nuclear norm regularized estimation of panel regression models
- Statistical inference in tensor completion: optimal uncertainty quantification and statistical-to-computational gaps
- A Unified Framework for Estimation of High-Dimensional Conditional Factor Models
- Inference in unbalanced panel data models with interactive fixed effects
- Learning non-parametric basis independent models from point queries via low-rank methods
- Inference on low-rank data matrices with applications to microarray data
This page was built for publication: Inference for low-rank models
Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q6177325)
