An algorithm for the multivariate group lasso with covariance estimation

I. Wilms*, C. Croux

*Corresponding author for this work

Research output: Contribution to journalArticleAcademicpeer-review

Abstract

Abstract we study a group lasso estimator for the multivariate linear regression model that accounts for correlated error terms. A block coordinate descent algorithm is used to compute this estimator. We perform a simulation study with categorical data and multivariate time series data, typical settings with a natural grouping among the predictor variables. Our simulation studies show the good performance of the proposed group lasso estimator compared to alternative estimators. We illustrate the method on a time series data set of gene expressions.
Original languageEnglish
Pages (from-to)668-681
Number of pages14
JournalJournal of Applied Statistics
Volume45
Issue number4
DOIs
Publication statusPublished - 2018

Keywords

  • Categorical variables
  • group lasso
  • multivariate regression
  • penalized maximum likelihood
  • sparsity
  • time series
  • SPARSE-GROUP LASSO
  • REGRESSION
  • SELECTION
  • MODEL

Cite this