Ensuring identifiability in hierarchical mixed effects Bayesian models.
Ensuring identifiability in hierarchical mixed effects Bayesian models.
复制标题
确保分层混合效应贝叶斯模型的可识别性。
DOI：
--
复制
发表时间：
2020
影响因子：
5
通讯作者：
Jarrett J. Barber
中科院分区：
文献类型：
--
作者：
K. Ogle;Jarrett J. Barber
生态学家越来越熟悉贝叶斯统计建模及其相关的马尔可夫链蒙特卡罗（MCMC）方法，以推断或发现数据中有趣的影响。生态数据的复杂性往往意味着实施（统计）模型具有非常丰富的效应结构，包括交叉或嵌套（即，分级或多级）固定和/或随机效应结构。然而，我们的经验表明，大多数生态学家并不熟悉这些模型及其在流行软件中的实现经常出现的微妙但重要的问题。对我们来说，最重要的考虑因素是效应可识别性的概念，它通常涉及数据、模型或实施方法的信息质量，即，识别感兴趣的数量。在本文中，我们专注于实现陷阱，可能误导后续的推理，尽管否则翔实的数据和模型。我们使用合成数据的随机效应回归来说明上述问题。我们将展示如何诊断可识别性问题，以及如何在流行的软件中使用模型重新参数化和计算和/或编码实践来修复这些问题，重点是JAGS，OpenBUGS和Stan。我们还展示了如何将这些解决方案扩展到更复杂的模型，涉及多组嵌套，交叉，加性或乘法效应，涉及随机和/或固定效应的模型。最后，我们提供了示例代码（JAGS / OpenBUGS和Stan），从业者可以修改并用于自己的应用程序。
Ecologists are increasingly familiar with Bayesian statistical modeling and its associated Markov chain Monte Carlo (MCMC) methodology to infer about or to discover interesting effects in data. The complexity of ecological data often suggests implementation of (statistical) models with a commensurately rich structure of effects, including crossed or nested (i.e., hierarchical or multi-level) structures of fixed and/or random effects. Yet, our experience suggests that most ecologists are not familiar with subtle but important problems that often arise with such models and with their implementation in popular software. Of foremost consideration for us is the notion of effect identifiability, which generally concerns how well data, models, or implementation approaches inform about-i.e., identify-quantities of interest. In this paper, we focus on implementation pitfalls that potentially misinform subsequent inference, despite otherwise informative data and models. We illustrate the aforementioned issues using random effects regressions on synthetic data. We show how to diagnose identifiability issues and how to remediate these issues with model reparameterization and computational and/or coding practices in popular software, with a focus on JAGS, OpenBUGS, and Stan. We also show how these solutions can be extended to more complex models involving multiple groups of nested, crossed, additive, or multiplicative effects, for models involving random and/or fixed effects. Finally, we provide example code (JAGS / OpenBUGS and Stan) that practitioners can modify and use for their own applications.
登录
查看更多内容
影响因子：
5.8
作者：
Carpenter, Bob;Gelman, Andrew;Riddell, Allen
通讯作者：
Riddell, Allen
影响因子：
3.4
作者：
Lemoine, Nathan P.
通讯作者：
Lemoine, Nathan P.
影响因子：
5.1
作者：
Michener, William K.
通讯作者：
Michener, William K.
影响因子：
2
作者：
Eberly, LE;Carlin, BP
通讯作者：
Carlin, BP