Ústav teorie informace a automatizace

Jste zde

Bibliografie

Journal Article

On selection of optimal stochastic model for accelerated life testing

Volf Petr, Timková Jana

: Reliability Engineering & System Safety vol.131, 1 (2014), p. 291-297

: Reliability analysis, Goodness-of-fit, Bayes statistics

: 10.1016/j.ress.2014.04.015

: http://library.utia.cas.cz/separaty/2015/SI/volf-0444111.pdf

(eng): This paper deals with the problem of proper lifetime model selection in the context of statistical reliability analysis. Namely, we consider regression models describing the dependence of failure intensities on a covariate, for instance a stressor. Testing the model fit is standardly based on the so-called martingale residuals. Their analysis has already been studied by many authors. Nevertheless, the Bayes approach to the problem is just developing. We shall present the Bayes procedure of stimation in several semi-parametric regression models of failure intensity. Then, our main concern is the Bayes construction of residual processes and goodness-of-fit tests basedon them.

: BB

07.01.2019 - 08:39