Transposable regularized covariance models with an application to missing data imputation
From MaRDI portal
Abstract: Missing data estimation is an important challenge with high-dimensional data arranged in the form of a matrix. Typically this data matrix is transposable, meaning that either the rows, columns or both can be treated as features. To model transposable data, we present a modification of the matrix-variate normal, the mean-restricted matrix-variate normal, in which the rows and columns each have a separate mean vector and covariance matrix. By placing additive penalties on the inverse covariance matrices of the rows and columns, these so-called transposable regularized covariance models allow for maximum likelihood estimation of the mean and nonsingular covariance matrices. Using these models, we formulate EM-type algorithms for missing data imputation in both the multivariate and transposable frameworks. We present theoretical results exploiting the structure of our transposable models that allow these models and imputation methods to be applied to high-dimensional data. Simulations and results on microarray data and the Netflix data show that these imputation techniques often outperform existing methods and offer a greater degree of flexibility.
Recommendations
- Missing values: sparse inverse covariance estimation and an extension to sparse regression
- An Imputation–Regularized Optimization Algorithm for High Dimensional Missing Data Problems and Beyond
- High-dimensional covariance matrix estimation with missing observations
- Pattern alternating maximization algorithm for missing data in high-dimensional problems
- Minimax rate-optimal estimation of high-dimensional covariance matrices with incomplete data
Cites work
- scientific article; zbMATH DE number 4159863 (Why is no real title available?)
- scientific article; zbMATH DE number 1834445 (Why is no real title available?)
- scientific article; zbMATH DE number 1391247 (Why is no real title available?)
- Are a set of microarrays independent of each other?
- Choosing starting values for the EM algorithm for getting the highest likelihood in multivariate Gaussian mixture models
- Component selection and smoothing in multivariate nonparametric regression
- Correlated \(z\)-values and the accuracy of large-scale statistical estimates
- Covariance-regularized regression and classification for high dimensional problems
- Exact matrix completion via convex optimization
- Maximum likelihood estimation via the ECM algorithm: A general framework
- Multiple Imputation After 18+ Years
- Sparse inverse covariance estimation with the graphical lasso
- Sparse permutation invariant covariance estimation
- Stochastic versions of the em algorithm: an experimental study in the mixture case
- The mle algorithm for the matrix normal distribution
- Transposable regularized covariance models with an application to missing data imputation
- Variable Selection via Nonconcave Penalized Likelihood and its Oracle Properties
Cited in
(41)- A Penalized Likelihood Method for Classification With Matrix-Valued Predictors
- Kronecker-structured covariance models for multiway data
- Existence and uniqueness of the Kronecker covariance MLE
- Co-clustering of spatially resolved transcriptomic data
- Model selection and estimation in the matrix normal graphical model
- Detecting column dependence when rows are correlated and estimating the strength of the row correlation
- An expectation-maximization algorithm for the matrix normal distribution with an application in remote sensing
- Gaussian and robust Kronecker product covariance estimation: existence and uniqueness
- scientific article; zbMATH DE number 7626713 (Why is no real title available?)
- Transposable regularized covariance models with an application to missing data imputation
- Separable factor analysis with applications to mortality data
- Imputation and low-rank estimation with missing not at random data
- PLS for Big Data: a unified parallel algorithm for regularised group PLS
- A generalized least-square matrix decomposition
- Graphical model selection and estimation for high dimensional tensor data
- Mixture of multivariate Gaussian processes for classification of irregularly sampled satellite image time-series
- A constrained matrix-variate Gaussian process for transposable data
- Missing values: sparse inverse covariance estimation and an extension to sparse regression
- Mixed Hölder matrix discovery via wavelet shrinkage and Calderón-Zygmund decompositions
- Gemini: graph estimation with matrix variate normal instances
- Recovering networks from distance data
- Maximum likelihood estimation for matrix normal models via quiver representations
- Concentration of measure bounds for matrix-variate data with missing values
- Autoregressive identification of Kronecker graphical models
- Matrix Completion, Counterfactuals, and Factor Analysis of Missing Data
- Testing the mean matrix in high-dimensional transposable data
- Hypothesis Testing of Matrix Graph Model with Application to Brain Connectivity Analysis
- Existence and uniqueness of the maximum likelihood estimator for models with a Kronecker product covariance structure
- Sparse Matrix Graphical Models
- Estimating high-dimensional covariance and precision matrices under general missing dependence
- The mixed Lipschitz space and its dual for tree metrics
- Hypothesis testing for the covariance matrix in high-dimensional transposable data with Kronecker product dependence structure
- Scalable Bayesian matrix normal graphical models for brain functional networks
- Sampling, denoising and compression of matrices by coherent matrix organization
- Covariance estimation via sparse Kronecker structures
- Many-sample tests for the equality and the proportionality hypotheses between large covariance matrices
- Consistency of large dimensional sample covariance matrix under weak dependence
- Pattern alternating maximization algorithm for missing data in high-dimensional problems
- Inferring Phenotypic Trait Evolution on Large Trees With Many Incomplete Measurements
- Testing high-dimensional mean vector with applications. A normal reference approach
- Unifying and generalizing methods for removing unwanted variation based on negative controls
This page was built for publication: Transposable regularized covariance models with an application to missing data imputation
Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q993250)
