Paper Title
Statistical Estimation for the Difference of Variances in Delta-Lognormal Distributions

Abstract
The delta-lognormal distribution represents the data that contain both zero and positive observations. The variance is a measure of dispersion, whereas the square root of the variance is called a standard deviation. The goal of this paper was to present the generalized confidence interval based on Jeffreys (GCI-J) and the method of variance estimates discovery (MOVER) to construct confidence intervals for the difference between two variances of delta-lognormal distributions. The coverage probabilities and relative average length are evaluated the performance of presented confidence intervals via Monte Carlo simulation. Our findings concluded that the GCI performance satisfied the target even small sample, so it can be considered as the recommended method. Furthermore, MOVER is also recommended for large variance. All confidence intervals are utilized to analyze the real-world datasets in several fields, including the natural rainfall amount for agriculture and the distance traveled of mice for biology. Keywords - Biology, GCI, MOVER, Rainfall.