A Brief History of Generative Models for Power Law and Lognormal Distributions
M. Mitzenmacher. Internet Mathematics, 1 (2):
226--251(2004)
Abstract
Recently, I became interested in a current debate over whether file size distributions are best modelled by a power law distribution or a lognormal distribution. In trying to learn enough about these distributions to settle the question, I found a rich and long history, spanning many fields. Indeed, several recently proposed models from the computer science community have antecedents in work from decades ago. Here, I briefly survey some of this history, focusing on underlying generative models that
lead to these distributions. One finding is that lognormal and power law distributions connect quite naturally, and hence, it is not surprising that lognormal distributions have arisen as a possible alternative to power law distributions across many fields.
%0 Journal Article
%1 mitzenmacher2004history
%A Mitzenmacher, M.
%D 2004
%J Internet Mathematics
%K generative model statistics powerlaw
%N 2
%P 226--251
%T A Brief History of Generative Models for Power Law and Lognormal Distributions
%U http://www.eecs.harvard.edu/~michaelm/CS223/powerlaw.pdf
%V 1
%X Recently, I became interested in a current debate over whether file size distributions are best modelled by a power law distribution or a lognormal distribution. In trying to learn enough about these distributions to settle the question, I found a rich and long history, spanning many fields. Indeed, several recently proposed models from the computer science community have antecedents in work from decades ago. Here, I briefly survey some of this history, focusing on underlying generative models that
lead to these distributions. One finding is that lognormal and power law distributions connect quite naturally, and hence, it is not surprising that lognormal distributions have arisen as a possible alternative to power law distributions across many fields.
@article{mitzenmacher2004history,
abstract = {Recently, I became interested in a current debate over whether file size distributions are best modelled by a power law distribution or a lognormal distribution. In trying to learn enough about these distributions to settle the question, I found a rich and long history, spanning many fields. Indeed, several recently proposed models from the computer science community have antecedents in work from decades ago. Here, I briefly survey some of this history, focusing on underlying generative models that
lead to these distributions. One finding is that lognormal and power law distributions connect quite naturally, and hence, it is not surprising that lognormal distributions have arisen as a possible alternative to power law distributions across many fields.
},
added-at = {2011-12-21T14:09:46.000+0100},
author = {Mitzenmacher, M.},
biburl = {https://www.bibsonomy.org/bibtex/2acdeb6b7980b25477665939c191f1e40/jaeschke},
interhash = {50b0caa36c6cbc1ecfa0714157f06bd1},
intrahash = {acdeb6b7980b25477665939c191f1e40},
journal = {Internet Mathematics},
keywords = {generative model statistics powerlaw},
number = 2,
pages = {226--251},
timestamp = {2021-08-11T13:41:00.000+0200},
title = {A Brief History of Generative Models for Power Law and Lognormal Distributions
},
url = {http://www.eecs.harvard.edu/~michaelm/CS223/powerlaw.pdf},
volume = 1,
year = 2004
}