Estimation, prediction and interpretation of NGG random effects models: an application to Kevlar fibre failure times

Abstract

AperTO -Archivio Istituzionale Open Access dell'Università di TorinoEstimation, prediction and interpretation of NGG random effects models: An application to Kevlar fibre failure times / Argiento, R.; Pievatolo, A.. -55(2014), pp. 805-826. Original Citation:Estimation, prediction and interpretation of NGG random effects models: An application to Kevlar fibre failure times Compared to a previous parametric Bayesian analysis, we obtain narrower credibility intervals and a better fit to the data. We also fit a similar semiparametric model, which can be seen a a special case of ours, where the error is a scale mixture of Weibull densities, mixed by a Dirichlet process, whereas the shape parameter has a parametric prior. The adequacy of the two semiparametric models is comparable, but the more general model provides a different evaluation of the posterior variance of the left tail of the lifetime distribution, which is an effect of assuming a nonparametric prior on both parameters of the Weibull density.

Optimal stratification of survival data via Bayesian nonparametric mixtures

“…. , n. A second option, along similar lines to Argiento et al (2014) and named here model M1, consists in assuming the effect of the covariates is common to all individuals across strata, that is θ i = θ for all i = 1, . .…”

“…Additionally, available information on the number of strata can be incorporated into the model by tuning the prior distribution for the number of strata. Our contribution adds to an existing body of literature on nonparametric mixtures of accelerated life models for data analysis and clustering (Argiento et al, 2009(Argiento et al, , 2010(Argiento et al, , 2014Liverani et al, 2020).…”

“…Alternatively, nonparametric mixtures based on the use of mixing completely random measures have been deployed to model hazard functions (Dykstra and Laud, 1981;Nieto-Barajas and Walker, 2004), possibly accounting for the presence of covariates (see, e.g., Nieto- Barajas and Walker, 2005;Nipoti et al, 2018). Nonparametric mixtures for accelerated life models in the presence of covariates have been considered in Argiento et al (2014) and Liverani et al (2020), with the goal of producing cluster-specific posterior inference. The first reference -see also Argiento et al (2009Argiento et al ( , 2010 for allied work-proposes the definition of Weibull accelerated life models with mixing normalized generalized gamma process and models the effect of predictors as common across clusters; the second one models survival data with a DP mixture with generic kernel admitting a convenient factorization into the product of a response and a covariate model.…”