Shrinking characteristics of precision matrix estimators

DOI10.1093/BIOMET/ASY023MaRDI QIDQ4561011zbMATH OpenOpenAlexFDO

Authors Aaron J. Molstad, Adam J. Rothman

Publication date 10 December 2018

Published in Biometrika (Search for Journal in Brave)

Full work available at URL https://arxiv.org/abs/1704.04820

zbMATH Keywords

alternating direction method of multiplierslinear discriminant analysismajorize-minimizeprecision matrix estimationprediction

Mathematics Subject Classification ID

Estimation in multivariate analysis (62H12) Classification and discrimination; cluster analysis (statistical aspects) (62H30) Ridge regression; shrinkage estimators (Lasso) (62J07)

Abstract: We propose a framework to shrink a user-specified characteristic of a precision matrix estimator that is needed to fit a predictive model. Estimators in our framework minimize the Gaussian negative loglikelihood plus an $L_1$ penalty on a linear or affine function evaluated at the optimization variable corresponding to the precision matrix. We establish convergence rate bounds for these estimators and propose an alternating direction method of multipliers algorithm for their computation. Our simulation studies show that our estimators can perform better than competitors when they are used to fit predictive models. In particular, we illustrate cases where our precision matrix estimators perform worse at estimating the population precision matrix but better at prediction.

Recommendations

• Direct shrinkage estimation of large dimensional precision matrix
• scientific article; zbMATH DE number 7168259
• Shrinkage estimation of large dimensional precision matrix using random matrix theory
• Estimating sparse precision matrix: optimal rates of convergence and adaptive estimation
• Shrinkage tuning parameter selection in precision matrices estimation

Cited in

(8)

• Graph informed sliced inverse regression
• A covariance-enhanced approach to multitissue joint eQTL mapping with application to transcriptome-wide association studies
• Shrinkage estimation of large dimensional precision matrix using random matrix theory
• scientific article; zbMATH DE number 7255095 (Why is no real title available?)
• Fast and Separable Estimation in High-Dimensional Tensor Gaussian Graphical Models
• scientific article; zbMATH DE number 7168259 (Why is no real title available?)
• Learning the hub graphical Lasso model with the structured sparsity via an efficient algorithm
• Kernelized Discriminant Analysis for Joint Modeling of Multivariate Categorical Responses

This page was built for publication: Shrinking characteristics of precision matrix estimators

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q4561011)