Asymptotic properties of a double penalized maximum likelihood estimator in logistic regression

Sujuan Gao, Jianzhao Shen

Research output: Contribution to journalArticle

12 Scopus citations

Abstract

Maximum likelihood estimates in logistic regression may encounter serious bias or even non-existence with many covariates or with highly correlated covariates. In this paper, we show that a double penalized maximum likelihood estimator is asymptotically consistent in large samples.

Original languageEnglish (US)
Pages (from-to)925-930
Number of pages6
JournalStatistics and Probability Letters
Volume77
Issue number9
DOIs
StatePublished - May 1 2007

    Fingerprint

Keywords

  • Logistic regression
  • Maximum likelihood
  • Penalized maximum likelihood
  • Ridge regression

ASJC Scopus subject areas

  • Statistics, Probability and Uncertainty
  • Statistics and Probability

Cite this