A comprehensive study of lognormality tests

Authors

  • Mahdi Mahdizadeh Department of Statistics, Hakim Sabzevari University, Sabzevar, Iran
  • Ehsan Zamanzade Department of Statistics, University of Isfahan, Isfahan, Iran

DOI:

https://doi.org/10.1285/i20705948v10n2p349

Keywords:

Empirical distribution function, Information theory, Likelihood ratio.

Abstract

There is abundant and increasing evidence that the lognormal distribution can account for random variation present in the data from many scientific fields. In the light of this flexibility for modeling, this article deals with goodness-of-fit tests for the lognormal distribution. Several testing procedures are compared by means of extensive simulation. Lastly, an actuarial data set is analyzed for illustration.

Author Biographies

Mahdi Mahdizadeh, Department of Statistics, Hakim Sabzevari University, Sabzevar, Iran

Assistant Professor

Ehsan Zamanzade, Department of Statistics, University of Isfahan, Isfahan, Iran

Assistant Professor

Downloads

Published

14-10-2017

Issue

Vol. 10 No. 2 (2017): Electronic Journal of Applied Statistical Analysis

Section

Original Paper

License

Authors who publish with EJASA agree to the  Creative Commons Attribuzione - Non commerciale - Non opere derivate 3.0 Italia License.