Bayesian methods

贝叶斯方法

• 批准号: 8553987
• 负责人: Zhen Chen
• 金额: $ 8.15万
• 依托单位: EUNICE KENNEDY SHRIVER NATIONAL INSTITUTE OF CHILD HEALTH & HUMAN DEVELOPMENT
• 依托单位国家: 美国
• 资助国家: 美国
• 起止时间: 至
• 项目状态: 未结题
• 来源: https://reporter.nih.gov/project-details/8553987
• 关键词: Agreement; Bayesian Method; Complex; Computational algorithm; Data; Data Analyses; Diagnostic; Epidemiology; Markov Chains; Measurement; Methods; Modeling; Monte Carlo Method; National Institute of Child Health and Human Development; Pregnancy; Time; statistics

Bayesian methods in epidemiology has advanced rapidly in the past two decades. Yet a lot need to be done to accommodate new challenges arising from expanded and new epidemiology data. This PI initiated project focuses on building modeling machinery for time to pregnancy data and for data on measurement agreement and diagnostic statistics. Moreover, the project also investigates new Bayesian computational algorithm for handling massive epidemiology data using sequential Markov chain Monte Carlo method or variational approach.

流行病学中的贝叶斯方法在过去二十年中发展迅速。然而，需要做很多工作，以适应扩大和新的流行病学数据带来的新挑战。该PI发起的项目侧重于构建妊娠时间数据和测量一致性和诊断统计数据的建模机制。此外，本计划亦研究新的贝氏演算法，以序贯马尔可夫链蒙地卡罗法或变分法处理大量流行病学资料。