The Korean Journal of Applied Statistics (응용통계연구)

The Korean Statistical Society (한국통계학회)
권호 표지
DOI QR코드

DOI QR Code

On Prediction Intervals for Binomial Data

이항자료에 대한 예측구간

  • Ryu, Jea-Bok (Department of Statistics, College of Science & Engineering, Cheongju University)
  • 류제복 (청주대학교 이공대학 통계학과)
  • Received : 2013.08.26
  • Accepted : 2013.10.24
  • Published : 2013.12.31
Download PDF
⟨ Previous Next ⟩

Abstract

Wald, Agresti-Coull, Jeffreys, and Bayes-Laplace methods are commonly used for confidence interval of binomial proportion are applied for prediction intervals. We used coverage probability, mean coverage probability, root mean squared error, and mean expected width for numerical comparisons. From the comparisons, we found that Wald is not proper as for confidence interval and Agresti-Coull is too conservative to differ from confidence interval. However, Jeffrey and Bayes-Laplace are good for prediction interval and Jeffrey is especially desirable as for confidence interval.

신뢰구간 추정에 널리 사용되고 있는 Wald, Agresti-Coull, 그리고 베이지안 방법인 Jeffrey와 Bayes-Laplace를 예측구간에 적용하였다. 네 가지 방법의 수치적 비교를 위해서 포함확률, 평균포함확률, 평균제곱오차의 제곱근, 그리고 평균기대폭을 사용하였다. 비교결과 Wald 방법은 신뢰구간에서와 마찬가지로 예측구간에서도 바람직하지 않았고 신뢰구간에서 선호되던 Agresti-Coull 방법은 예측구간에서는 너무 보수적이라 적절치 않다. 반면에 Jeffrey와 Bayes-Laplace 방법은 적절하였고, 특히 Jeffrey 방법은 신뢰구간의 경우에서와 마찬가지로 예측구간에서도 바람직하였다.

Keywords

References

  1. Agresti, A. and Coull, B. A. (1998). Approximate is better than "Exact" for interval estimation of Binomial proportions, The American Statistician, 52, 119-126.
  2. Brown, L. D., Cai, T. T. and DasGupta, A. (2001). Interval estimation for a binomial proportion(with discussion), Statistical Science, 16, 101-133.
  3. Brown, L. D., Cai, T. T. and DasGupta, A. (2002). Confidence intervals for a binomial proportion and asymptotic expansions, The Annals of Statistics, 30, 160-201. https://doi.org/10.1214/aos/1015362189
  4. Geisser, S. (1984). On prior distribution for binary trials(with discussion), The American Statistician, 38, 244-247.
  5. Hahn, G. J. and Meeker, W. Q. (1991). Statistical Intervals - A Guide for Practitioners, John Wiley & Sons, Inc.
  6. Hall, P. and Rieck, A. (2001). Improving coverage accuracy of nonparametric prediction intervals, Journal of the Royal Statistical Society, Series B, 63, 717-725. https://doi.org/10.1111/1467-9868.00308
  7. Nelson, W. B. (2004). Applied Life Data Analysis, John Wiley & Sons, Inc.
  8. Ryu, J. B. (2009). A short consideration of binomial confidence interval, Communications of the Korean Statistical Society, 16, 731-743. https://doi.org/10.5351/CKSS.2009.16.5.731
  9. Ryu, J. B. (2010). The effect of adjusting the extreme values inWald confidence interval, Journal of Research Institute of Industrial Sciences, 28, 29-34.
  10. Ryu, J. B. (2011). The influence of extreme value in binomial confidence interval, Communications of the Korean Statistical Society, 18, 615-623. https://doi.org/10.5351/CKSS.2011.18.5.615
  11. Ryu, J. B. and Lee, S. J. (2006). Confidence intervals for a low binomial proportion, The Korean Journal of Applied Statistics, 19, 217-230. https://doi.org/10.5351/KJAS.2006.19.2.217
  12. Tuyl, F., Gerlach, R. and Mengersen, K. (2008). A comparison of Bayes-Laplace, Jeffreys, and other priors: The case of zero events, The American Statistician, 62, 40-44. https://doi.org/10.1198/000313008X267839
  13. Tuyl, F., Gerlach, R. and Mengersen, K. (2009). Posterior predictive arguments in favor of the Bayes-Laplace prior as the consensus priors for binomial and multinomial parameters, Bayesian Analysis, 4, 151-158. https://doi.org/10.1214/09-BA405
  14. Vidoni, P. (2009). Improving prediction intervals and distributional functions, Scandinavian Journal of Statistics, 36, 735-748. https://doi.org/10.1111/j.1467-9469.2009.00656.x
  15. Wang, H. (2008). Coverage probability of prediction intervals for discrete random variable, Computational Statistics and Data Analysis, 53, 17-26. https://doi.org/10.1016/j.csda.2008.07.017
  16. Winkler, R. L., Smith, J. E. and Fryback, D. G. (2002). The role of informative priors in zero-numerator problems: Being conservative versus being candid, The American Statistician, 56, 1-4. https://doi.org/10.1198/000313002753631295
  17. Yu, K. and Ally, A. (2009). Improving prediction intervals: Some elementary methods, The American Statistician, 63, 17-19. https://doi.org/10.1198/tast.2009.0003