|
Activity Number:
|
178
|
|
Type:
|
Topic Contributed
|
|
Date/Time:
|
Monday, August 4, 2008 : 2:00 PM to 3:50 PM
|
|
Sponsor:
|
Section on Bayesian Statistical Science
|
| Abstract - #302388 |
|
Title:
|
Incorporating Structural Equation Modeling to Fecundity Models
|
|
Author(s):
|
Sung Duk Kim*+ and Rajeshwari Sundaram and Germaine B. Louis
|
|
Companies:
|
National Institute of Child Health and Human Development and National Institutes of Health and National Institute of Child Health and Human Development
|
|
Address:
|
6100 Executive Boulevard, Rockville, MD, 20852,
|
|
Keywords:
|
Conception ; Fertility ; Latent variables ; Markov chain Monte Carlo ; Menstrual Cycle ; Structural Equations Model
|
|
Abstract:
|
In epidemiologic studies, human fecundability is measured by probability of conception in a menstrual cycle by a couple. We propose a general model for joint modeling of intercourse behavior and human fecundability through a classic conception probability model and structural equation model which involves a set of latent variables to capture dependence between intercourse behavior on consecutive days in a menstrual cycle, so that the proposed model can accommodate not only a broad variety of intercourse patterns and dependency structure, but also general covariate effects, heterogeneity among fecund couples in menstrual cycle viability and in frequency of intercourse. Markov Chain Monte Carlo method is used to carry out Bayesian posterior computation. A detailed analysis of the New York State Angler pregnancy data is presented to illustrate the proposed methodology.
|
