PREDICTING FUTURE DATA FROM GAMMA-MIXTURE AND BETA-MIXTURE DISTRIBUTIONS AND APPLICATION TO THE RECOVERY RATE OF COVID-19

Abstract

This paper addresses the problem of predicting future order statistics arising from mixture distributions. This study generalizes Barakat et al.’s [1] work that focused only on the data from a mixture of exponential distributions to wide models including the mixture-gamma and beta-mixture distributions. For these two lifetime models, we construct prediction intervals and point predictions for future observations. Two cases are considered, in the first case, we assume a fixed sample size, while in the second case, the sample size is assumed to be a positive integer-valued random variable independent of the observations. A simulation study is carried out for corroborating the finding and illustrative purposes. Finally, a practical application of the proposed method is applied to the recovery rate of COVID-19 data sets. Received: August 3, 2023Accepted: September 18, 2023