Likelihood adaptively modified penalties

Yang Feng, Tengfei Li, Zhiliang Ying

Research output: Contribution to journalArticlepeer-review

Abstract

A new family of penalty functions, ie, adaptive to likelihood, is introduced for model selection in general regression models. It arises naturally through assuming certain types of prior distribution on the regression parameters. To study the stability properties of the penalized maximum-likelihood estimator, 2 types of asymptotic stability are defined. Theoretical properties, including the parameter estimation consistency, model selection consistency, and asymptotic stability, are established under suitable regularity conditions. An efficient coordinate-descent algorithm is proposed. Simulation results and real data analysis show that the proposed approach has competitive performance in comparison with the existing methods.

Original languageEnglish (US)
Pages (from-to)330-353
Number of pages24
JournalApplied Stochastic Models in Business and Industry
Volume35
Issue number2
DOIs
StatePublished - Mar 1 2019

Keywords

  • Poisson penalty
  • asymptotic stability
  • consistency
  • convexity
  • negative absolute prior
  • penalized likelihood
  • sigmoid penalty

ASJC Scopus subject areas

  • Modeling and Simulation
  • Business, Management and Accounting(all)
  • Management Science and Operations Research

Fingerprint Dive into the research topics of 'Likelihood adaptively modified penalties'. Together they form a unique fingerprint.

Cite this