Dimension asymptotics for generalised bootstrap in linear regression

Research output: Contribution to journalArticle

4 Scopus citations

Abstract

The consistency of a class of generalized bootstrap techniques for the distribution of the least squares parameter estimator in linear regression was examined. The data size and the regressors were assumed to be random in the analysis. The results of the analysis showed that best results were obtained with resampling techniques assuming that the number of parameters tend to infinity.

Original languageEnglish (US)
Pages (from-to)367-381
Number of pages15
JournalAnnals of the Institute of Statistical Mathematics
Volume54
Issue number2
DOIs
StatePublished - Oct 9 2002

    Fingerprint

Keywords

  • Bootstrap
  • Dimension asymptotics
  • Jackknife
  • Regression

Cite this