The Counter-intuitive Non-informative Prior for the Bernoulli Family


Authors: 
Zhu, M., & Lu, A. Y.
Category: 
Volume: 
12(2)
Pages: 
Online
Year: 
2004
Publisher: 
Journal of Statistics Education
URL: 
http://www.amstat.org/publications/jse/v12n2/zhu.pdf
Abstract: 

In Bayesian statistics, the choice of the prior distribution is often controversial. Different rules for selecting priors have been suggested in the literature, which, sometimes, produce priors that are difficult for the students to understand intuitively. In this article, we use a simple heuristic to illustrate to the students the rather counter-intuitive fact that flat priors are not necessarily non-informative; and non-informative priors are not necessarily flat.

The CAUSE Research Group is supported in part by a member initiative grant from the American Statistical Association’s Section on Statistics and Data Science Education

register