Latent diffusion models for event history analysis

Keywords

Statistical learning
Code:
22/2007
Title:
Latent diffusion models for event history analysis
Date:
Tuesday 20th November 2007
Author(s):
Roberts, Gareth O.; Sangalli, Laura M.
Download link:
Abstract:
We consider Bayesian hierarchical models for event history analysis, where the event times are modeled through an underlying diffusion process, which determines the hazard rate. We show how these models can be efficiently treated by means of Markov chain Monte Carlo techniques.
This report, or a modified version of it, has been also submitted to, or published on
Gareth O. Roberts and Laura M. Sangalli (2010), Latent diffusion models for survival analysis, Bernoulli, Vol. 16, pp. 435-458.