Conditional Inference on the Generalized Gamma Distribution Parameters

Abstract

Abstract It is widely known that the Bayes method is usually efficient especially when the sample sizes are small or when the data are heavily censored. However, the subjectivity of the Bayesian method to the prior distribution can take a posterior decision to unwelcome consequences. Therefore, this paper's main objective is to find the conditional point estimates using the pivotal functions. The conditional inference was used to estimate the generalized gamma distribution parameters, based on the generalized progressive hybrid-censoring scheme, and compare it with the Bayesian estimates, via Monte Carlo simulation. The simulation results showed that the conditional inference is highly efficient and provides better estimates than the Bayesian estimates based on different loss functions. Finally, from a future perspective, the proposed model could be important for analysing biostatistics data on COVID-19 deaths in Egypt to demonstrate the efficiencies of the proposed methods. AMS Subject Classification: 62F15; 62F40; 62F86; 62N01; 62N02