Using OLS to test for normality

Research output: Contribution to journalArticlepeer-review

6 Scopus citations

Abstract

The OLS estimator is a weighted average of the slopes delineated by adjacent observations. These weights depend only on the independent variable. Equal weights are obtained if and only if the independent variable is normally distributed. This feature is used to develop a new test for normality which is compared to standard tests and provides better power for testing normality.

Original languageEnglish
Pages (from-to)2050-2058
Number of pages9
JournalStatistics and Probability Letters
Volume82
Issue number11
DOIs
StatePublished - 1 Nov 2012

Keywords

  • Jarque-Bera test
  • Kolmogorov-Smirnov test
  • Regression weights
  • Shapiro-Wilk test

Fingerprint

Dive into the research topics of 'Using OLS to test for normality'. Together they form a unique fingerprint.

Cite this