Confidence Intervals for the Ratio of Coefficients of Variation in the Two-Parameter Exponential Distributions

摘要

Ratio of coefficients of variation is one of statistical measurements used in many fields of applied research. However, the problem in statistical inference of the ratio of coefficients of variation has been little studied. In this paper, two new confidence intervals for this measure in the two-parameter exponential distributions are introduced based on the method of variance of estimates recovery (MOVER) and the generalized confidence interval (GCI). We use Monte carlo simulation to conduct the performance of the estimators. The results indicate that the coverage probabilities of the confidence interval based on the MOVER and the GCI maintain the nominal coverage level. The GCI has shorter expected length than the MOVER in most cases. In addition, real-world data are analyzed to illustrate the findings of the paper.