A Bayesian Hypothesis Testing Procedure Possessing the Concept of Significance Level

  • Hwang, Hyungtae (Professeor, Department of Computer Science and Statistics, Dankook University, Seoul, 140-714)
  • 발행 : 2001.12.01

초록

In this paper, Bayesian hypothesis testing procedures are proposed under the non-informative prior distributions, which can be thought as the Bayesian counterparts of the classical ones in the sense of using the concept of significance level. The performances of proposed procedures are compared with those of classical procedures through several examples.

키워드

참고문헌

  1. Statistical Decision theory and Bayesian Bayesian analysis (Second edition) Berger,J.
  2. Theoretical Statistics Cox,D.R.;Hinkley,D.V.
  3. The Korean Communications in Statistics v.8 no.1 A Bayesian criterion for a multiple test of two mulitvariate normal population Kim,H.J.;Son,Y.S.
  4. Bayesian Statistics : an Introduction Lee,P.M.
  5. Testing Statistical Hypotheses (Second edition) Lehmann,E.L.
  6. Biometrika v.44 A statistical paradox Lindley,D.V.
  7. Introductio to Probability and Statistics from a Bayesian Viewpoint Lindley,D.V.
  8. An Introduction to Bayesian Inference in Econometrics Zellner,A.