Villa, Cristiano, Walker, Stephen G. (2017) On The Mathematics of The JeffreysLindley Paradox. Communications in Statistics – Theory and Methods, 46 (24). pp. 1229012298. ISSN 03610926. (doi:10.1080/03610926.2017.1295073)
PDF  Author's Accepted Manuscript  
Download (202kB)
Preview



Official URL http://dx.doi.org/10.1080/03610926.2017.1295073 
Abstract
This paper is concerned with the well known JeffreysLindley paradox. In a Bayesian set up, the socalled paradox arises when a point null hypothesis is tested and an objective prior is sought for the alternative hypothesis. In particular, the posterior for the null hypothesis tends to one when the uncertainty, i.e. the variance, for the parameter value goes to infinity. We argue that the appropriate way to deal with the paradox is to use simple mathematics, and that any philosophical argument is to be regarded as irrelevant.
Item Type:  Article 

DOI/Identification number:  10.1080/03610926.2017.1295073 
Uncontrolled keywords:  Bayes factor, Bayesian hypothesis testing, Kullback–Leibler divergence, selfinformation loss 
Subjects:  Q Science > QA Mathematics (inc Computing science) > QA276 Mathematical statistics 
Divisions:  Faculties > Sciences > School of Mathematics Statistics and Actuarial Science > Statistics 
Depositing User:  Cristiano Villa 
Date Deposited:  08 Feb 2017 08:37 UTC 
Last Modified:  11 Jul 2019 13:39 UTC 
Resource URI:  https://kar.kent.ac.uk/id/eprint/60238 (The current URI for this page, for reference purposes) 
Villa, Cristiano:  https://orcid.org/0000000226702954 
 Export to:
 RefWorks
 EPrints3 XML
 BibTeX
 CSV
 Depositors only (login required):