Comparing particulate matter dispersion in Thailand using the Bayesian Confidence Intervals for ratio of coefficients of variation

Recently, harmful levels of air pollution have been detected in many provinces of Thailand. Particulate matter (PM) contains microscopic solids or liquid droplets that are so small that they can be inhaled and cause serious health problems. A high dispersion of PM is measured by a coefficient of variation of log-normal distribution. Since the log-normal distribution is often used to analyse environmental data such as hazardous dust particle levels and daily rainfall data. These data focus the statistical inference on the coefficient of variation. In this paper, we develop confidence interval estimation for the ratio of coefficients of variation of two log-normal distributions constructed using the Bayesian approach. These confidence intervals were then compared with the existing approaches: method of variance estimates recovery (MOVER), modified MOVER, and approximate fiducial approaches using their coverage probabilities and average lengths via Monte Carlo simulation. The simulation results show that the Bayesian confidence interval performed better than the others in terms of coverage probability and average length. The proposed approach and the existing approaches are illustrated using examples from data set PM10 level and PM2.5 level in the northern Thailand.