EM Algorithm for Mixture Distributions Model with Type-I Hybrid Censoring Scheme

An expectation–maximization (EM) likelihood estimation procedure is proposed to obtain the maximum estimates of parameters in a mixture distributions model based on type-I hybrid censored samples when proportions are unknown. Three bootstrap methods applied construct confidence intervals parameters. Monte Carlo simulations conducted evaluate performance methods. Simulation results show that can perform well reliable point and interval results. examples used illustrate applications