Burr type XII distribution (BXIID) has earned more attention in the few past decades because of the flexibility of distribution shape for model fitting. However, no analytical closed formula solutions of the maximum likelihood estimates (MLEs) of BXIID parameters can be obtained based on progressively type I interval-censored (PTIIC) samples. In this manuscript, the differential evolution algorithm (DE) and quasi-Newton method (QN) are applied to searching the MLEs of BXIID parameters based on PTIIC samples. The performance of DE and QN is evaluated by means of Monte Carlo simulations. Simulation results show that the DE outperforms QN in terms of smaller bias and mean squared error (MSE) of the MLEs.
Parameter estimation for burr type XII distribution with.... Available from: https://www.researchgate.net/publication/319494371_Parameter_estimation_for_burr_type_XII_distribution_with_differential_evolution_and_Quasi-Newton_approaches_based_on_progressively_type_I_interval-censored_samples [accessed May 09 2018].