A semi-parametric mixture model in fertility studies |
| |
| Institution: | 1. Department of Biostatistics, University of North Carolina at Chapel Hill, Chapel Hill, NC 27599-7400, USA;2. Biostatistics Branch, MD A3-03, National Institute of Environmental Health Sciences, P.O. Box 12233, Research Triangle Park NC 27709, USA;1. Graduate School of Health Sciences, Gunma University, 3-39-22 Showa-machi, Maebashi, Gunma 371-8511, Japan;2. Center for Food Science and Wellness, Gunma University, 4-2 Aramaki, Maebashi, Gunma 371-8510, Japan;1. Division of Allergy and Immunology, Department of Pediatrics, UNC School of Medicine, Chapel Hill, NC;2. Marsico Lung Institute, University of North Carolina at Chapel Hill, Chapel Hill, NC;3. Department of Biostatistics, University of North Carolina at Chapel Hill, Chapel Hill, NC;4. Campbell University College of Pharmacy and Health Sciences, Buies Creek, NC;1. Ecole Centrale de Nantes - CNRS, 1 rue de la Noe, 44300 Nantes, France;2. SBM Offshore, 24 avenue de Fontvieille, MC98007, Monaco |
| |
| Abstract: | We describe a semiparametric mixture model for human fertility studies. The probability of conception is a product of two components. The mixing distribution, the component that introduces the heterogeneity among the menstrual cycles that come from different couples, is characterized nonparametrically by a finite number of moments. The second component, the intercourse-related probability is modeled parametrically to assess the possible exposure effects. We discuss an EM algorithm-based estimating procedure that incorporates the natural order in the moments. |
| |
| Keywords: | |
| 本文献已被 ScienceDirect 等数据库收录！ |
|
