pvalues for model evaluation
Abstract
Deciding whether a model provides a good description of data is often based on a goodnessoffit criterion summarized by a pvalue. Although there is considerable confusion concerning the meaning of pvalues, leading to their misuse, they are nevertheless of practical importance in common data analysis tasks. We motivate their application using a Bayesian argumentation. We then describe commonly and less commonly known discrepancy variables and how they are used to define pvalues. The distribution of these are then extracted for examples modeled on typical data analysis tasks, and comments on their usefulness for determining goodnessoffit are given.
 Publication:

Physical Review D
 Pub Date:
 January 2011
 DOI:
 10.1103/PhysRevD.83.012004
 arXiv:
 arXiv:1011.1674
 Bibcode:
 2011PhRvD..83a2004B
 Keywords:

 02.50.r;
 Probability theory stochastic processes and statistics;
 Physics  Data Analysis;
 Statistics and Probability
 EPrint:
 33 pages, 13 figures