Fixing Bias in Zipf's Law Estimators Using Approximate Bayesian Computation

08/03/2020

∙

by   Charlie Pilgrim, et al.

∙

0

∙

share

The prevailing Bayesian maximum likelihood estimators for inferring power law models from rank-frequency data are biased. The source of this bias is an inappropriate likelihood function. The correct likelihood function is derived and shown to be computationally intractable. A more computationally efficient method of approximate Bayesian computation is described that estimates Zipf exponents for large datasets without bias.