응용통계연구 (The Korean Journal of Applied Statistics)

  • 제7권2호
  • /
  • Pages.9-19
  • /
  • 1994
  • /
  • 1225-066X(pISSN)
  • /
  • 2383-5818(eISSN)
한국통계학회 (The Korean Statistical Society)
권호 표지

베타-이항 분포에서 Gibbs sampler를 이용한 평가 일치도의 사후 분포 추정

Posterior density estimation of Kappa via Gibbs sampler in the beta-binomial model

  • 엄종석 ((136-042) 서울시 성북구 삼선동 2가 한성대학교 전산통계학과) ;
  • 최일수 ((120-749) 서울시 서대문구 신촌동 연세대학교 응용통계학과) ;
  • 안윤기 ((120-749) 서울시 서대문구 신촌동 연세대학교 응용통계학과)
  • 발행 : 1994.09.01
PDF 다운로드
⟨ 이전 논문 다음 논문 ⟩

초록

평가자간 평가 일치도(measure of agreement)를 나타내는 모수 $\kappa$와 양성 반응 비율 $\mu$를 지닌 베타-이항 분포 모형은 심리학 분야에서 많이 다루어지는 모형이다. 이 모형에서 $\kappa$에 대한 추정은 $\mu$가 0에 가까운 값을 가질 때 우도함수를 이용한 전통적 추론 방법의 적용이 어렵다. 본 논문에서는 이러한 문제를 Gibbs sampler를 이용한 Bayesian 분석 방법을 적용시켜 주변 사후 밀도 함수를 추정하였으며 이를 이용하여 Bayesian 추정값도 구하였다.

Beta-binomial model, which is reparametrized in terms of the mean probability $\mu$ of a positive deagnosis and the $\kappa$ of agreement, is widely used in psychology. When $\mu$ is close to 0, inference about $\kappa$ become difficult because likelihood function becomes constant. We consider Bayesian approach in this case. To apply Bayesian analysis, Gibbs sampler is used to overcome difficulties in integration. Marginal posterior density functions are estimated and Bayesian estimates are derived by using Gibbs sampler and compare the results with the one obtained by using numerical integration.

키워드

참고문헌

  1. Statistical Decision Theory and Bayesian Analysis Berger,J.O.
  2. Statistics and Computing v.1 An Iterative Monte Carlo Method for Nonconjugate Bayesian Analysis Carlin,B.P.;Gelfand,A.E.
  3. Journal of the Royal Stotistical Society. Ser.C. v.41 Hierachical Bayesian Analysis of Change Point Problems Carlin,B.P.;Gelfand,A.E.;Smith,A.F.M.
  4. Educational and Psychological Measuremet v.20 A Coefficient of Agreement for Nominal Scales Cohen,J.
  5. Journal of the American Statistical Association v.85 Sampling-based Approaches to Calculating Marginal Densities Gelfand,A.E.;Smith,A.F.M.
  6. IEEE, PAMI-6 v.6 Stochastic Relaxation, Gibbs distributions, and the Bayesian Restoration of Images Geman,S.;Geman,D.
  7. Mutation research v.41 The Distribution of Fetal Death in Control Mice and its Implications on Statistical Test for Dominant Lethal Effects Haseman;Soares
  8. Biometrics v.34 The use of a Correlated Binomial Model for the analysis of certain Toxicological Experiments Haseman;Kupper
  9. Econometrica v.46 Bayesian estimates of equation system parameters; an application of integration by Monte Carlo Kloeck,T.;van Dijk,H.K.
  10. Psychometrika v.44 Ramifications of a population model for ĸ as a coefficient of reliability Kraemer,H.C.
  11. Approximate Bayesian Methods, in Bayesian Statistics Lindley,D.V.;J.M.Bernardo(eds.);M.H.Degroot(eds.);D.V.Lindley(eds.);A.M.F.Smith(eds.)
  12. Journal of Chemical Physics v.21 Equation of State Calculations by Fast Computer Machines Metropolis,N.;Rosenbluth,A.W.;Rosenbluth,M.N.;Teller,A.H.;Teller,E.
  13. The American Statistician v.46 Bayesian statistics without tears: A sampling-resampling perspective Smith,A.F.M.;Gelfand,A.E.
  14. Journal of the American Statistical Association v.82 The calculation of posterior distributions by data augmentation Tanner,M.A.;Wong,W.H.
  15. Journal of the American Statistical Association v.81 Accurate Approximations for Posterior Moments and Marginal Densities Tierney,L.;Kadane,J.B.
  16. Ph.D Thesis Conditional inference about kappa in the Beta-binomial model Um,J.S.
  17. Journal of the American Statistical Association v.85 Posterior Computations with Censored Regression Data Wei,G.C.G.;Tanner,M.A.
  18. Journal of the American Statistical Association v.86 Generalized Linear Models with Random Effects: A Gibbs Sampling Approach Zeger,S.;Karim,M.R.