Simultaneous Confidence Intervals for All Differences of Coefficients of Variation of Two-Parameter Exponential Distributions

Warisa  Thangjai; Suparat  Niwitpong

Warisa Thangjai Department of Statistics, Faculty of Science, Ramkhamhaeng University, Bangkok, Thailand
Suparat Niwitpong Department of Applied Statistics, Faculty of Applied Science, King Mongkut’s University of Technology North Bangkok, Bangkok, Thailand

Keywords: Coverage probability, simulation, parametric bootstrap approach, generalized confidence interval approach, method of variance estimates recovery approach

Abstract

In this paper, a construction of simultaneous confidence intervals for differences of coefficients of variation of k independent two-parameter exponential distributions is considered. Parametric bootstrap approach, generalized confidence interval (GCI) approach, and method of variance estimates recovery (MOVER) approach are proposed to construct the simultaneous confidence intervals. Simulation results indicated that the GCI approach has satisfactory performance for all cases whereas the MOVER approach is preferable for large sample sizes.

Downloads

Download data is not yet available.