Regression to the mean for nonnormal populations

Kenneth J. Beath*, Annette J. Dobson

*Corresponding author for this work

Research output: Contribution to journalArticlepeer-review

2 Citations (Scopus)

Abstract

Regression to the mean is the term used to describe the effect by which individuals selected on the basis of an initial measurement which is particularly high or low compared with other members of the population have, upon remeasurement, values nearer to the population mean. For the normal distribution simple expressions for estimates of the magnitude of this effect are available. Many biological variables, however, are not normally distributed. Estimates for regression to the mean for nonnormal data are given based on Edgeworth series and saddlepoint approximations.

Original languageEnglish
Pages (from-to)431-435
Number of pages5
JournalBiometrika
Volume78
Issue number2
DOIs
Publication statusPublished - Jun 1991
Externally publishedYes

Keywords

  • Edgeworth series
  • Regression to the mean
  • Saddlepoint approximation

Fingerprint

Dive into the research topics of 'Regression to the mean for nonnormal populations'. Together they form a unique fingerprint.

Cite this