Extensions to the W-Test for Normality.

Abstract

A method of testing for normality is given, based on a linear model for the expected value of an order statistic. It is an extension of the technique of Shapiro and Wilk, and is related to earlier ideas of Neyman and Barton. Several alternative test procedures are considered, with the same asymptotic theory for the test statistics; this theory is developed, and some results are given, for n finite, for one of the procedures. LaBrecque has recently developed another of the procedures and the connections are shown.

Document Details

Document Type
Technical Report
Publication Date
Jan 31, 1975
Accession Number
ADA013084

Entities

People

  • M. A. Stephens

Organizations

  • Stanford University

Tags

DTIC Thesaurus Topics

  • Computing-Related Activities
  • Data Science
  • Information Science
  • Interdisciplinary Science
  • Mathematics
  • Normality
  • Statistics

Fields of Study

  • Mathematics

Readers

  • Statistical inference.