Ústav teorie informace a automatizace

Jste zde

Bibliografie

Journal Article

Non-linear failure rate: A Bayes study using Hamiltonian Monte Carlo simulation

Volf Petr, Thach T., Briš R., Coolen F.

: International Journal of Approximate Reasoning vol.123, 1 (2020), p. 55-76

: Non-linear failure rate, Bayesian estimators, Hamiltonian Monte Carlo

: 10.1016/j.ijar.2020.04.007

: http://library.utia.cas.cz/separaty/2020/SI/volf-0524681.pdf

: https://www.sciencedirect.com/science/article/pii/S0888613X20301596

(eng): A non-linear failure ratemodel is introduced, analyzed, and applied to real data sets for both censored and uncensored data. The Hamiltonian Monte Carlo and cross-entropy methods have been exploited to empower the traditional methods of statistical estimation. Bayes estimators of parameters and reliability characteristics uses the Hamiltonian Monte Carlo and these estimators are considered under both symmetric and asymmetric loss functions. Additionally, the maximum likelihood estimators of parameters are obtained by using the cross-entropy method to optimize the log-likelihood function. The superiority of the proposed model and estimation procedures are demonstrated on real data sets.

: BB

: 10103

07.01.2019 - 08:39