Title :
Testing the statistical similarity of discrete observations using Dirichlet priors
Author :
Lynch, Robert S., Jr. ; Willett, Peter K.
Author_Institution :
Naval Undersea Warefare Center, Newport, RI, USA
Abstract :
In this paper the combined Bayes test, which uses a Dirichlet (noninformative) prior, is used to determine if the underlying statistics of two data populations are the same. Results demonstrate that the combined Bayes test performs better than the more conventional Kolmogorov-Smirnov test
Keywords :
Bayes methods; error statistics; information theory; testing; Bayes test; Dirichlet priors; Kolmogorov-Smirnov test; data populations; discrete observations; error probability; statistical similarity; testing; Bayesian methods; Contracts; Degradation; Histograms; Integrated circuit testing; Laboratories; Performance evaluation; Probability; Quantization; Statistical analysis;
Conference_Titel :
Information Theory, 1998. Proceedings. 1998 IEEE International Symposium on
Conference_Location :
Cambridge, MA
Print_ISBN :
0-7803-5000-6
DOI :
10.1109/ISIT.1998.708736
