This article reviews simulation methods for the beta-Stacy process, which belongs to the family of neutral-to-the-right (NTR) processes used in Bayesian nonparametric inference on survival distributions. The definition and properties of the NTR processes are recalled, with a view to represent the continuous component of the process at the integrated hazard level via the Lévy measure. The conditional distribution of the prior process, given right-censored data, is described; this represents the key ingredient for devising sampling schemes. Four methods aiming at the generation of increments or trajectories of the hazard rate process are revised. We explain the ideas behind each of them and sketch the correspondent algorithms for the posterior beta-Stacy process.
Simulation of the Beta-Stacy process
DE BLASI, Pierpaolo
2007-01-01
Abstract
This article reviews simulation methods for the beta-Stacy process, which belongs to the family of neutral-to-the-right (NTR) processes used in Bayesian nonparametric inference on survival distributions. The definition and properties of the NTR processes are recalled, with a view to represent the continuous component of the process at the integrated hazard level via the Lévy measure. The conditional distribution of the prior process, given right-censored data, is described; this represents the key ingredient for devising sampling schemes. Four methods aiming at the generation of increments or trajectories of the hazard rate process are revised. We explain the ideas behind each of them and sketch the correspondent algorithms for the posterior beta-Stacy process.File | Dimensione | Formato | |
---|---|---|---|
deblasi-2007.pdf
Accesso riservato
Tipo di file:
POSTPRINT (VERSIONE FINALE DELL’AUTORE)
Dimensione
97.48 kB
Formato
Adobe PDF
|
97.48 kB | Adobe PDF | Visualizza/Apri Richiedi una copia |
I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.