Journal of the Korean Data and Information Science Society

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

Asymptotic Inference on the Odds Ratio via Saddlepoint Method

안부점근사를 이용한 승산비에 대한 점근적 추론

  • Na, Jong-Hwa (Dept. of Statistics, Chungbuk National University of Cheong-Ju)
  • Published : 1999.04.30
Download PDF
⟨ Previous Next ⟩

Abstract

We propose a new method of asymptotic inference on the odds ratio (or cross-product ratio) in $2{\times}2$ contingency table. Saddlepoint approximations to the conditional tail probability we used in this procedure. We assess the accuracy of the suggested method by comparing with the exact one. To obtain the exact values, we need very complicated calculations containing the cumulative probabilities of non-central hypergeometric distribution. The suggested method in this paper is very accurate even for small or moderate sample sizes as well as simple and easy to use. Example with a real data is also considered.

분할표 분석에서 승산비 (odds ratio)에 대한 추론은 중요하다. 이에 대한 정확한 추론은 비중심초기하(noncentral hypergeometric) 분포의 누적확률등의 계산이 요구되어 표본의 크기가 클 경우 많은 양의 계산과 계산시간이 요구되므로 StatXact 등의 프로그램을 이용하는 것이 일반적이다. 본 논문에서는 정확한 추론에 대한 대안적 방법으로 안부점 근사(saddlepoint approximation)의 결과를 이용한 점근적 추론법을 제시하였다. 이 방법은 비교적 소표본의 경우에도 정확한 추론의 결과와 일치하며, 기존의 정규근사를 이용한 방법에 비해 매우 뛰어난 정확도를 유지함을 예제를 통해 확인하였다.

Keywords

References

  1. Biometrika v.73 no.2 Inference on full or partial partial parameters based on the standardized signed log-likelihood ratio Barndorff-Nislesen, O. E.
  2. Journal of the Royal Statistical Society (Series B) v.41 no.3 Edgeworth and saddlepoint approximations with statistical applications Barndorff-Nielsen, O. E.;Cox, D. R.
  3. International Statistical Review v.55 no.1 Tail probability approximations Daniels, H. E.
  4. Proceedings of the Royal Statistical Society of London(Series A) v.144 Two new properties of mathematical likelihood Fisher, R. A.
  5. Australian Journal of Statistics v.4 Confidence limits for a cross-product ratio Fisher, R. A.
  6. Biometrika v.77 Tail probabilities from observed likelihood Fraser, D. A. S.
  7. Journal of the American Statistical Association v.86 no.414 Statistical inference: Likelihood to significance Fraser, D. A. S.
  8. Journal of the Royal Statistical Society (Series B) v.54 no.3 Practical use of higher order asymptotics for multiparameter exponential families Pierce, D. A.;Peters, D.
  9. Journal of Applied Probability v.24 Saddlepoint expansions for conditional distributions Skovgaard, I. M.