Commentaries: Use of the correlation coefficient with normal probability plots

Stephen W. Looney, Thomas R. Gulledge

Research output: Contribution to journalComment/debate

123 Citations (Scopus)

Abstract

The use of the correlation coefficient is suggested as a technique for summarizing and objectively evaluating the information contained in probability plots. Goodness-of-fit tests are constructed using this technique for several commonly used plotting positions for the normal distribution. Empirical sampling methods are used to construct the null distribution for these tests, which are then compared on the basis of power against certain nonnormal alternatives. Commonly used regression tests of fit are also included in the comparisons. The results indicate that use of the plotting position pi = (i – .375)/(n + .25) yields a competitive regression test of fit for normality.

Original languageEnglish (US)
Pages (from-to)75-79
Number of pages5
JournalAmerican Statistician
Volume39
Issue number1
DOIs
StatePublished - Feb 1985

Fingerprint

Probability Plot
Correlation coefficient
Regression
Null Distribution
Goodness of Fit Test
Sampling Methods
Pi
Normality
Gaussian distribution
Alternatives

Keywords

  • Empirical power comparison
  • Filliben test
  • Plotting position
  • Regression tests of fit
  • Shapiro-Francia test
  • Shapiro-Wilk test

ASJC Scopus subject areas

  • Statistics and Probability
  • Mathematics(all)
  • Statistics, Probability and Uncertainty

Cite this

Commentaries : Use of the correlation coefficient with normal probability plots. / Looney, Stephen W.; Gulledge, Thomas R.

In: American Statistician, Vol. 39, No. 1, 02.1985, p. 75-79.

Research output: Contribution to journalComment/debate

@article{d7f348493f7c4ecc92f8559e71a647cd,
title = "Commentaries: Use of the correlation coefficient with normal probability plots",
abstract = "The use of the correlation coefficient is suggested as a technique for summarizing and objectively evaluating the information contained in probability plots. Goodness-of-fit tests are constructed using this technique for several commonly used plotting positions for the normal distribution. Empirical sampling methods are used to construct the null distribution for these tests, which are then compared on the basis of power against certain nonnormal alternatives. Commonly used regression tests of fit are also included in the comparisons. The results indicate that use of the plotting position pi = (i – .375)/(n + .25) yields a competitive regression test of fit for normality.",
keywords = "Empirical power comparison, Filliben test, Plotting position, Regression tests of fit, Shapiro-Francia test, Shapiro-Wilk test",
author = "Looney, {Stephen W.} and Gulledge, {Thomas R.}",
year = "1985",
month = "2",
doi = "10.1080/00031305.1985.10479395",
language = "English (US)",
volume = "39",
pages = "75--79",
journal = "American Statistician",
issn = "0003-1305",
publisher = "American Statistical Association",
number = "1",

}

TY - JOUR

T1 - Commentaries

T2 - Use of the correlation coefficient with normal probability plots

AU - Looney, Stephen W.

AU - Gulledge, Thomas R.

PY - 1985/2

Y1 - 1985/2

N2 - The use of the correlation coefficient is suggested as a technique for summarizing and objectively evaluating the information contained in probability plots. Goodness-of-fit tests are constructed using this technique for several commonly used plotting positions for the normal distribution. Empirical sampling methods are used to construct the null distribution for these tests, which are then compared on the basis of power against certain nonnormal alternatives. Commonly used regression tests of fit are also included in the comparisons. The results indicate that use of the plotting position pi = (i – .375)/(n + .25) yields a competitive regression test of fit for normality.

AB - The use of the correlation coefficient is suggested as a technique for summarizing and objectively evaluating the information contained in probability plots. Goodness-of-fit tests are constructed using this technique for several commonly used plotting positions for the normal distribution. Empirical sampling methods are used to construct the null distribution for these tests, which are then compared on the basis of power against certain nonnormal alternatives. Commonly used regression tests of fit are also included in the comparisons. The results indicate that use of the plotting position pi = (i – .375)/(n + .25) yields a competitive regression test of fit for normality.

KW - Empirical power comparison

KW - Filliben test

KW - Plotting position

KW - Regression tests of fit

KW - Shapiro-Francia test

KW - Shapiro-Wilk test

UR - http://www.scopus.com/inward/record.url?scp=84952511519&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84952511519&partnerID=8YFLogxK

U2 - 10.1080/00031305.1985.10479395

DO - 10.1080/00031305.1985.10479395

M3 - Comment/debate

AN - SCOPUS:84952511519

VL - 39

SP - 75

EP - 79

JO - American Statistician

JF - American Statistician

SN - 0003-1305

IS - 1

ER -