Models for lifetime data in the presence of competing causes with different activation mechanisms

• Grant number: 11/22924-4
• Support Opportunities: Scholarships abroad - Research
• Start date: July 02, 2012
• End date: July 01, 2013
• Field of knowledge: Physical Sciences and Mathematics - Probability and Statistics - Statistics
• Principal Investigator: Vicente Garibay Cancho
• Grantee: Vicente Garibay Cancho
• Host Investigator: Dipak Kumar Dey
• Host Institution: Instituto de Ciências Matemáticas e de Computação (ICMC). Universidade de São Paulo (USP). São Carlos , SP, Brazil
• Institution abroad: University of Connecticut (UCONN), United States

Abstract

In this project we propose new models for survival data with and without a fraction of cure. The proposed models arise in a scenario of latent competitive causes (risks), where the number of causes (latent variable) for the occurrence of a particular event of interest is modelled by a power series distribution. For the proposed models, we develop inferential procedures from classical and Bayesian perspectives.Classical inferential approach is based on the likelihood method. From the Bayesian point of view however we explore the use of Monte Carlo methods (MCMC). (AU)