Sparse Reduced Rank Huber Regression in High Dimensions
From MaRDI portal
Abstract: We propose robust sparse reduced rank regression for analyzing large and complex high-dimensional data with heavy-tailed random noise. The proposed method is based on a convex relaxation of a rank- and sparsity-constrained non-convex optimization problem, which is then solved using the alternating direction method of multipliers algorithm. We establish non-asymptotic estimation error bounds under both Frobenius and nuclear norms in the high-dimensional setting. This is a major contribution over existing results in reduced rank regression, which mainly focus on rank selection and prediction consistency. Our theoretical results quantify the tradeoff between heavy-tailedness of the random noise and statistical bias. For random noise with bounded th moment with , the rate of convergence is a function of , and is slower than the sub-Gaussian-type deviation bounds; for random noise with bounded second moment, we obtain a rate of convergence as if sub-Gaussian noise were assumed. Furthermore, the transition between the two regimes is smooth. We illustrate the performance of the proposed method via extensive numerical studies and a data application.
Cites work
- scientific article; zbMATH DE number 1301873 (Why is no real title available?)
- A General Qualitative Definition of Robustness
- A general Bahadur representation of M-estimators and its application to linear regression with nonstochastic designs
- Adaptive Huber Regression
- Asymptotic behavior of M estimators of p regression parameters when \(p^ 2/n\) is large. II: Normal approximation
- Asymptotics with increasing dimension for robust regression with applications to the bootstrap
- Distributed optimization and statistical learning via the alternating direction method of multipliers
- Empirical properties of asset returns: stylized facts and statistical issues
- Guaranteed minimum-rank solutions of linear matrix equations via nuclear norm minimization
- I-LAMM for sparse learning: simultaneous control of algorithmic complexity and statistical error
- Joint variable and rank selection for parsimonious estimation of high-dimensional matrices
- On Cross-Validation for Sparse Reduced Rank Regression
- On the Douglas-Rachford splitting method and the proximal point algorithm for maximal monotone operators
- Optimal selection of reduced rank estimators of high-dimensional matrices
- Outlier detection using nonconvex penalized regression
- Reduced rank stochastic regression with a sparse singular value decomposition
- Reduced rank regression via adaptive nuclear norm penalization
- Reduced rank ridge regression and its kernel extensions
- Reduced-rank regression for the multivariate linear model
- Robust Estimation of a Location Parameter
- Robust principal component analysis?
- Robust reduced-rank regression
- Robust regression: Asymptotics, conjectures and Monte Carlo
- Selective factor extraction in high dimensions
- Signal extraction approach for sparse multivariate response regression
- Sparse reduced-rank regression for simultaneous dimension reduction and variable selection
- The cluster graphical Lasso for improved estimation of Gaussian graphical models
Cited in
(8)- Personalized federated learning on large-scale association networks
- Robust learning of the modified Huber loss
- Resistant convex clustering: how does the fusion penalty enhance resistance?
- Efficient and provable online reduced rank regression via online gradient descent
- Asymptotically faster estimation of high-dimensional additive models using subspace learning
- Nuclear norm regularization
- On Cross-Validation for Sparse Reduced Rank Regression
- Sparse reduced-rank regression with covariance estimation
This page was built for publication: Sparse Reduced Rank Huber Regression in High Dimensions
Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q6144754)
