Bayesian Computation for Mean of Zero-Inflated Rayleigh Distribution With Application to COVID-19 Data in Myanmar

We proposed the Bayesian computation for a mean of Zero-inflected Rayleigh distribution based on the Bayesian Markov chain Monte Carlo interval (MCMC) and Bayesian-Highest Posterior Density Interval (HPD) compared with the percentile bootstrap and the bootstrap confidence interval with standard error. The performance of the confidence interval estimators for zero-inflected rayleigh distribution were studied in terms of their coverage probability (CP) and expected length (EL) via Monte Carlo simulation by the R program. The results indicate the Bayesian MCMC and Bayesian HPD were more outstanding than or close to the nominal confidence level of 0.95. The expected length of Bayesian HPD is slightly smaller than the Bayesian MCMC. The simulation results show that the method recommended by this research for zero-inflected rayleigh distribution is Bayesian HPD. In empirical data, we used mortality rate data for COVID-19 from Myanmar to explain the performance of the proposed method.