Journal of the Korean Data and Information Science Society

The Korean Data and Information Science Society (한국데이터정보과학회)
권호 표지

Noninformative Priors for the Ratio of Means of Two Poisson Distributions

  • Published : 2002.10.31
Download PDF
⟨ Previous Next ⟩

Abstract

In this paper, Jeffrey's and reference priors are derived when the parameter of interest is the ratio of means of two in dependent Poisson distribution. To achieve the parameter orthogonality in the sense of Cox and Reid (1987), non-trivial orthogonal transformation is provided. The orthogonal transformation makes to find noninformative priors easy. Our simulation study indicates that the reference prior meet very well the target coverage probabilities in a frequentist sense. Using the real data, we compute Bayes estimator and MLE for the ratio of means based on the reference prior.

Keywords

References

  1. Journal of the American Statistical Association v.84 Estimating a Product of Means : Bayesian Analysis with Reference Priors Berger, J. O.;Bernardo, J. M.
  2. Bayesian Statistics IV On the Development of Reference Priors (with discussion) Berger, J. O.;Bernardo, J. M.
  3. Recent Developments in Statistics Inferences about the Ratio of Normal Means; A Bayesian Approach to the Fieller-Creasy Problem Bernardo, J. M.;J. R. Barra(et al.)(ed.)
  4. Journal of Royal Statistical Society v.B no.41 Reference Posterior Distributions for Bayesian Inference (with discussion) Bernardo, J. M.
  5. Journal of Royal Statistical Society v.B no.49 Orthogonal Parameters and Approximate Conditional Inference (with discussion) Cox, D. R.;Reid, N.
  6. Journal of the American Statiscal Association v.90 Some Remarks on Noninformative Priors Datta, G. S.;Ghosh, M.
  7. Biometrika v.80 Elimination of Nuisance Parameters with Reference Noninformative Priors Liseo, G. S.;Ghosh, M.
  8. Statistical Methods, Ames Snedecor, G. W.;Cochran, W. G.