Noppadon Yosboonruang.. Confidence intervals for coefficients of variation of the delta-lognormal distribution. Doctoral Degree(Applied Statistics). King Mongkut's University of Technology North Bangkok. Central Library. : King Mongkut's University of Technology North Bangkok, 2020.
Confidence intervals for coefficients of variation of the delta-lognormal distribution
Abstract:
The coefficient of variation is commonly used to measure the dispersion of data
and can be used to compare the degree of variation between two or more datasets
since it is independent of the units of measurement. In some study populations, data
are positive right-skewed observations with an excess of zeros, the distribution of
which is delta-lognormal. The purpose of this thesis is to construct confidence
intervals for the coefficients of variation and their functions of several delta-
lognormal distributions based on the concepts of the generalized confidence interval,
the fiducial generalized confidence interval, priors based on Bayesian methodology,
and the method of variance estimate recovery. In this thesis, the construction of
confidence intervals for the coefficient of variation of a delta-lognormal distribution is
first reported. Subsequently, the confidence intervals for the functions (the difference and the ratio) of two independent coefficients of variation of delta-lognormal distributions are proposed. Next, confidence intervals for the common coefficient of variation of delta-lognormal distribution are constructed. Following this, constructing simultaneous confidence intervals for all pairwise differences of the coefficients of variation of several delta-lognormal populations is presented. The results of verifying the performances of the proposed methods by analyzing their coverage probabilities together with their expected lengths via Monte Carlo simulation are then reported. Last, rainfall series from Thailand are used to illustrate the efficacy of the proposed confidence intervals.