Asymptotic confidence ellipses of parameters for the Birnbaum-Saunders distribution

Authors

  • Pattaya Thonglim Department of Mathematics and Statistics, Faculty of Science and Technology, Thammasat University, Patum Thani, Thailand
  • Kamon Budsaba Department of Mathematics and Statistics, Faculty of Science and Technology, Thammasat University, Patum Thani, Thailand.
  • Andrei I. Volodin Department of Mathematics and Statistics, University of Regina, Regina, Saskatchewan, Canada.

Keywords:

method of moment, method of maximum likelihood, confidence ellipse, monte carlo simulation

Abstract

The purpose of this study is to find the suitable covariance matrix for the construction of confidence regions of parameters in the Birnbaum-Saunders distribution and we need to calculate confidence ellipses and compare the coverage probabilities for asymptotic confidence ellipses of parameters in the Birnbaum-Saunders distribution. Monte Carlo simulation is used to compare the coverage probabilities of the asymptotic confidence ellipses. The result showed that the asymptotic confidence ellipses can work very well when the \alpha values increase more than 2.0 and the sample sizes (n) increase. In the Birnbaum-Saunders distribution, we can use method of moment estimators instead of maximum likelihood estimators for confidence ellipses because of high efficiency of coverage probabilities

Downloads

How to Cite

Thonglim, P., Budsaba, K., & Volodin, A. I. (2015). Asymptotic confidence ellipses of parameters for the Birnbaum-Saunders distribution. Thailand Statistician, 12(2), 207–222. Retrieved from https://ph02.tci-thaijo.org/index.php/thaistat/article/view/34200

Issue

Vol. 12 No. 2 (2014): JULY

Section

Articles