Bayesian Shrinkage for Functional Network Models, With Applications to Longitudinal Item Response Data

Bayesian Shrinkage for Functional Network Models, With Applications to Longitudinal Item Response Data
复制标题

功能网络模型的贝叶斯收缩及其在纵向项目响应数据中的应用

DOI:
10.1080/10618600.2021.1999823
复制
发表时间:
2022
影响因子:
2.4
通讯作者:
Jin, Ick Hoon
Jin, Ick Hoon
中科院分区:
数学2区
文献类型:
--
作者:
Park, Jaewoo;Jeon, Yeseul;Shin, Minsuk;Jeon, Minjeong;Jin, Ick Hoon

文献摘要

参考文献

相似文献

纵向项目反应数据在社会科学、教育科学和心理学等学科中很常见。研究项目之间随时间变化的关系对于教育评估或根据调查问题设计营销策略至关重要。虽然动态网络模型已经得到了广泛的发展,我们不能直接将它们应用到项目响应数据,因为有多个系统的节点与各种类型的项目之间的局部相互作用,导致多元化的网络结构。我们提出了一个新的模型来研究这些项目之间的时间相互作用,通过嵌入指数随机图模型框架内的功能参数。由于似然函数包含难以处理的归一化常数,因此对此类模型的推断是困难的。此外,函数参数的数量随着项目数量的增加而呈指数增长。这种模型的变量选择是不平凡的,因为标准的收缩方法不考虑功能参数的时间趋势。为了克服这些挑战,我们开发了一种新的贝叶斯方法相结合的辅助变量MCMC算法和最近开发的功能收缩方法。我们将我们的算法应用于调查和审查数据集,说明所提出的方法可以避免难以处理的归一化常数的评估,以及项目之间的显着的时间相互作用的检测。通过在不同场景下的仿真研究,我们检查我们的算法的性能。据我们所知,我们的方法是第一次尝试选择函数变量的模型与棘手的归一化常数。本文的补充材料可在网上查阅。
Longitudinal item response data are common in social science, educational science, and psychology, among other disciplines. Studying the time-varying relationships between items is crucial for educational assessment or designing marketing strategies from survey questions. Although dynamic network models have been widely developed, we cannot apply them directly to item response data because there are multiple systems of nodes with various types of local interactions among items, resulting in multiplex network structures. We propose a new model to study these temporal interactions among items by embedding the functional parameters within the exponential random graph model framework. Inference on such models is difficult because the likelihood functions contain intractable normalizing constants. Furthermore, the number of functional parameters grows exponentially as the number of items increases. Variable selection for such models is not trivial because standard shrinkage approaches do not consider temporal trends in functional parameters. To overcome these challenges, we develop a novel Bayes approach by combining an auxiliary variable MCMC algorithm and a recently developed functional shrinkage method. We apply our algorithm to survey and review datasets, illustrating that the proposed approach can avoid the evaluation of intractable normalizing constants as well as the detection of significant temporal interactions among items. Through a simulation study under different scenarios, we examine the performance of our algorithm. Our method is, to our knowledge, the first attempt to select functional variables for models with intractable normalizing constants. Supplementary materials for this article are available online.
DOI: 10.1080/10618600.2018.1448832
发表时间: 2018-01-01
影响因子: 2.4
作者:
Bouranis, Lampros;Friel, Nial;Maire, Florian
通讯作者: Maire, Florian
DOI: 10.1016/j.csda.2020.107052
发表时间: 2017-02
期刊: Comput. Stat. Data Anal.
影响因子: --
作者:
Jihui Lee;Gen Li;James D. Wilson
通讯作者: Jihui Lee;Gen Li;James D. Wilson
DOI: 10.1214/12-aos1044
发表时间: 2013-04
影响因子: 4.5
作者:
Shalizi CR;Rinaldo A
通讯作者: Rinaldo A
DOI: 10.1080/01621459.2019.1654875
发表时间: 2020
影响因子: 3.7
作者:
Shin M;Bhattachrya A;Johnson VE
通讯作者: Johnson VE
Bergm:R 中的贝叶斯指数随机图
DOI: 10.18637/jss.v061.i02
发表时间: 2012
影响因子: 5.8
作者:
A. Caimo;N. Friel
通讯作者: N. Friel