|
Activity Number:
|
325
|
|
Type:
|
Topic Contributed
|
|
Date/Time:
|
Tuesday, August 8, 2006 : 2:00 PM to 3:50 PM
|
|
Sponsor:
|
Section on Bayesian Statistical Science
|
| Abstract - #306475 |
|
Title:
|
A Comprehensive Spatial-Temporal Analysis of Breast Cancer: First Primary, Second Primary, and Breast Cancer Survival
|
|
Author(s):
|
Song Zhang*+
|
|
Companies:
|
M. D. Anderson Cancer Center
|
|
Address:
|
8450 Cambridge Street, Apt. 1175, Houston, TX, 77054,
|
|
Keywords:
|
breast cancer ; spatial-temporal analysis ; Bayesian p-splines ; hazard ; survival analysis
|
|
Abstract:
|
We propose a comprehensive spatial-temporal analysis of breast cancer, including first primary occurrence, second primary occurrence, and breast cancer survival. We assume a Poisson model for first primary incidence. We are interested in two competing risks: developing second primary occurrence and death from first primary. For each risk, we define a semiparametric age-hazard function that allows us to piece together information from subjects of different ages and construct a much longer hazard curve. The unknown baseline age-hazard curve is modeled by a Bayesian P-spline prior. Random spatial and temporal effects are assumed for the log-rates of the first primary occurrence and for the hazard functions. They are modeled by MCAR priors such that each component is allowed to have a different spatial-temporal pattern. A simulation study and real data analysis are presented.
|
- The address information is for the authors that have a + after their name.
- Authors who are presenting talks have a * after their name.
Back to the full JSM 2006 program |
