Two-Sample Prediction of Odd Generalized Exponential Inverted Weibull Distribution with theApplication on COVID-19 Mortality Rate

One of the most crucial issues in life testing is statistical prediction, which has also been used in business, engineering, medicine, and other fields. When more information are available, a better choice will be promoted. When projecting business results, prediction is used to save time, effort, and money. The predictor might be either a point predictor or an interval predictor. The main aim of this research is to investigate the two-sample prediction problem from the odd generalized exponential inverted Weibull distribution based on Type II censored samples. Furthermore, point and interval predictions for future order statistics using non-Bayesian, Bayesian, and E-Bayesian models are looked at. Future order statistics point and interval projections are also offered using conditional, maximum likelihood, Bayesian, and E-Bayesian techniques. The Bayesian and E-Bayesian predictors are based on two different loss functions: the balanced squared error loss function, which is symmetric, and the balanced linear exponential loss function, which is asymmetric. The predictors are derived using uniform hyper prior distributions and gamma prior distributions. Results have been applied to real data sets (such as the COVID-19 death rate in different countries) as well as simulation studies to show the flexibility and potential applications of the distribution.