Edgeworth expansions for M-estimators of a regression parameter

  • S. N. Lahiri

Research output: Contribution to journalArticlepeer-review

Abstract

This paper gives rth order Edgeworth expansions (r ≥ 3) for M-estimators of a regression parameter in a simple linear regression model. The regularity conditions used here essentially require the smoothness of some integrals of the score function with respect to the underlying error distribution. As a result, these expansions are valid for robust M-estimators corresponding to nonsmooth score functions.

Original languageEnglish
Pages (from-to)125-132
Number of pages8
JournalJournal of Multivariate Analysis
Volume43
Issue number1
DOIs
StatePublished - Oct 1992

Keywords

  • Edgeworth expansion
  • location model
  • M-estimator
  • regression

Fingerprint

Dive into the research topics of 'Edgeworth expansions for M-estimators of a regression parameter'. Together they form a unique fingerprint.

Cite this