Better models for Gibrat’s data

Saralees Nadarajah*, Emmanuel Afuecheta

*Corresponding author for this work

Research output: Contribution to journalArticlepeer-review


Akhundjanov and Toda (Empir Econ 59:2071–2091, 2020) analyzed 24 data sets in as reported by Gibrat (Les Inégalités Économiques, Librairie du Recueil Sirey, Paris, 1931), showing among others that 17 of the data sets can be best modeled by the Pareto-lognormal distribution due to Reed and Jorgensen (Commun Stat Theory Methods 33:1733–1753, 2004). Here, we reanalyze the same data sets and show that a new distribution exhibiting polynomial tails can provide even better fits. The assessment of better fits is based on two information criteria and likelihood ratio tests.

Original languageEnglish
Pages (from-to)2057-2067
Number of pages11
JournalEmpirical Economics
Issue number4
StatePublished - Apr 2022

Bibliographical note

Funding Information:
The authors would like to thank the Editor and the two referees for careful reading and comments which greatly improved the paper.

Publisher Copyright:
© 2021, The Author(s), under exclusive licence to Springer-Verlag GmbH Germany, part of Springer Nature.


  • Estimation
  • Polynomial tails
  • Student’s t distribution

ASJC Scopus subject areas

  • Statistics and Probability
  • Mathematics (miscellaneous)
  • Social Sciences (miscellaneous)
  • Economics and Econometrics


Dive into the research topics of 'Better models for Gibrat’s data'. Together they form a unique fingerprint.

Cite this