夏志明Zhiming Xia

教授

 博士生导师  硕士生导师
性别:男
学历:博士研究生毕业
在职信息:在职
所在单位:数学学院
电子邮箱:

论文成果

当前位置: 中文主页 >> 论文成果

Bayesian hierarchical model for analyzing multiresponse longitudinal pharmacokinetic data

点击次数:

影响因子:2.497
DOI码:10.1002/sim.7505
发表刊物:Statistics in Medicine
关键字:hierarchical model; MCMC algorithm; pharmacokinetic; traditional Chinese medicine
摘要:Traditional Chinese medicine (TCM) is a very complex mixture containing many different ingredients. Thus, statistical analysis of traditional Chinese medicine data becomes challenging, as one needs to handle the association among the observed data across different time points and across different ingredients of the multivariate response. This paper builds a 3-stage Bayesian hierarchical model for analyzing multivariate response pharmacokinetic data. Usually, the dimensionality of the parameter space is very huge, which leads to the parameter-estimation difficulty. So we take the hybrid Markov chain Monte Carlo algorithms to obtain the posterior Bayesian estimation of corresponding parameters in our model. Both simulation study and real-data analysis show that our theoretical model and Markov chain Monte Carlo algorithms perform well, and especially the correlation among different ingredients can be calculated very accurately.
论文类型:期刊论文
卷号:36
期号:30
页面范围:4816-4830
ISSN号:0277-6715
是否译文:
收录刊物:SCI
第一作者:Liping Zhao
通讯作者:Zhiming Xia