Scholarlink

<p>This paper introduces a linear mixed-effects model with an autoregressive correlation structure. This model includes a variable of interest with a threshold value such that the measurements below or above that cannot be quantified. The model employs Student’s-<i>t</i> distribution for the error terms and random effects, and a Markov chain Monte Carlo (MCMC) algorithm was developed to obtain Bayesian posterior distributions of unknown quantities of interest. The marginal likelihood function is used to compute the Bayesian model selection measures. The comparison between the proposed model and alternative models was discussed via synthetic data. Finally, the proposed methods were applied to longitudinal human immunodeficiency virus (HIV) viral load data from an acquired immunodeficiency syndrome (AIDS) study.</p>

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

Autoregressive Bayesian modeling of censored HIV longitudinal data using the multivariate Student’s-t distribution

  • 期刊论文

  • 发表日期：

    2025年3月23日

查看全文
  • Kelin Zhong,
  • Luis M. Castro,
  • Panpan Zhang,
  • Víctor H. Lachos

摘要

This paper introduces a linear mixed-effects model with an autoregressive correlation structure. This model includes a variable of interest with a threshold value such that the measurements below or above that cannot be quantified. The model employs Student’s-t distribution for the error terms and random effects, and a Markov chain Monte Carlo (MCMC) algorithm was developed to obtain Bayesian posterior distributions of unknown quantities of interest. The marginal likelihood function is used to compute the Bayesian model selection measures. The comparison between the proposed model and alternative models was discussed via synthetic data. Finally, the proposed methods were applied to longitudinal human immunodeficiency virus (HIV) viral load data from an acquired immunodeficiency syndrome (AIDS) study.