BFF: Bayesian, Fiducial, Frequentist Analysis of Age Effects in Daily Diary Data
BFF: Bayesian, Fiducial, Frequentist Analysis of Age Effects in Daily Diary Data
复制标题
BFF：每日日记数据中年龄影响的贝叶斯、基准、频率分析
DOI：
10.1093/geronb/gbz100
复制
发表时间：
2019
期刊：
影响因子：
--
通讯作者：
Ram, ed., Nilam
中科院分区：
文献类型：
--
作者：
Neupert, Shevaun D.;Hannig, Jan;Ram, ed., Nilam
ObjectivesWe应用新的统计模型每日日记数据推进方法和概念的目标。我们研究了每日日记数据中人内斜率的年龄效应，并引入了广义置信推断（GFI），它提供了频率论和贝叶斯推断之间的折衷。我们使用六个领域的日常压力暴露数据来生成具有日常负面影响的人内情绪反应斜率。我们测试系统的年龄差异和相似性，这些反应斜率，这是不一致的，在以前的research.Method101老年人（60-90岁）和108年轻（18-36岁）的成年人每天的压力和负面影响的问题，连续八天，导致1,438天。日常的压力域包括参数，避免参数，工作/志愿者的压力，家庭压力，网络压力，和健康相关的stressers.ResultsUsing贝叶斯，GFI，和频率主义的范式，我们比较了结果的6个压力域的重点解释年龄的影响在人内的反应。多水平模型表明，在避免争论，工作/志愿者压力，家庭压力和健康相关的压力领域内的每个范式的情绪反应的年龄效应为零。然而，这些模型在对争论和网络压力源的情绪反应中的零年龄效应方面存在分歧。讨论三种范式在六个压力源领域中的四个的反应中的零年龄效应上趋于一致。GFI是一个有用的工具，它提供了额外的信息时，作出决定，关于零年龄的影响，在人内斜率。我们为读者提供代码，以将这些模型应用于他们自己的数据。
ObjectivesWe apply new statistical models to daily diary data to advance both methodological and conceptual goals. We examine age effects in within-person slopes in daily diary data and introduce Generalized Fiducial Inference (GFI), which provides a compromise between frequentist and Bayesian inference. We use daily stressor exposure data across six domains to generate within-person emotional reactivity slopes with daily negative affect. We test for systematic age differences and similarities in these reactivity slopes, which are inconsistent in previous research.MethodOne hundred and eleven older (aged 60–90) and 108 younger (aged 18–36) adults responded to daily stressor and negative affect questions each day for eight consecutive days, resulting in 1,438 total days. Daily stressor domains included arguments, avoided arguments, work/volunteer stressors, home stressors, network stressors, and health-related stressors.ResultsUsing Bayesian, GFI, and frequentist paradigms, we compared results for the six stressor domains with a focus on interpreting age effects in within-person reactivity. Multilevel models suggested null age effects in emotional reactivity across each of the paradigms within the domains of avoided arguments, work/volunteer stressors, home stressors, and health-related stressors. However, the models diverged with respect to null age effects in emotional reactivity to arguments and network stressors.DiscussionThe three paradigms converged on null age effects in reactivity for four of the six stressor domains. GFI is a useful tool that provides additional information when making determinations regarding null age effects in within-person slopes. We provide the code for readers to apply these models to their own data.