Application of Bayesian Methods in Multilevel and Logitudinal Mediation Models

贝叶斯方法在多层次纵向中介模型中的应用

• 批准号: 7752931
• 负责人: Davood Tofighi
• 金额: $ 3.8万
• 依托单位: ARIZONA STATE UNIVERSITY-TEMPE CAMPUS
• 依托单位国家: 美国
• 财政年份: 2009
• 资助国家: 美国
• 起止时间: 2009-07-27 至 2011-07-26
• 项目状态: 已结题
• 来源: https://reporter.nih.gov/project-details/7752931
• 关键词: Accounting; Alcohol or Other Drugs use; Area; Bayesian Method; Complex; Computer software; Computers; Confidence Intervals; Data; Data Analyses; Data Set; Development; Drug usage; Education; Faculty; Fellowship; Foundations; Funding; Goals; Health Sciences; Intervention; Investigation; Joints; Lead; Longitudinal Studies; Markov Chains; Maximum Likelihood Estimate; Measurement; Measures; Mediating; Mediation; Methodology; Methods; Modeling; Monte Carlo Method; National Institute of Drug Abuse; Normal Statistical Distribution; Outcome; Outcome Measure; Paper; Participant; Performance; Pharmaceutical Preparations; Preparation; Prevention; Prevention Research; Prevention approach; Prevention program; Preventive Intervention; Process; Psychology; Randomized; Randomized Controlled Clinical Trials; Randomized Controlled Trials; Reading; Research; Research Design; Research Methodology; Research Personnel; Sample Size; Sampling; Solutions; Statistical Methods; Structure; Techniques; Testing; Time; Training; Work; Writing; base; career; computer code; data structure; design; improved; interest; member; multilevel analysis; peer; professor; psychosocial; public health relevance; research study; simulation; statistics; substance abuse prevention; theories; treatment effect

DESCRIPTION (provided by applicant): The goal of my proposed research is to develop Bayesian techniques to test mediational effects in cluster randomized designs with multiple measurement waves. Mediational analysis studies the processes through which interventions achieve their effects through intervening variables that are targeted for change. Because the widely used maximum likelihood approach relies on large samples and normal theory to produce valid results, Bayesian techniques are expected to show superior performance in small to moderate sample sizes particularly with the non-normal data common in substance abuse prevention trials. In addition, Bayesian methods can incorporate information from previous studies further adding to their efficiency. Aim 1 will develop Bayesian techniques to test mediation effects in cluster randomized and longitudinal models that can incorporate information from previous experiments and handle non-normal data. The resulting estimates can potentially lead to more reliable estimates and greater statistical power than current approaches. Aim 2 develops Markov Chain Monte Carlo (MCMC) methods to estimate Bayesian mediation models. Also, computer code to implement MCMC methods in publicly free software packages (e.g., R, WinBUGS) will be written, making this work accessible to other researchers. Aim 3 conducts a simulation study to compare the performance of the Bayesian and maximum likelihood estimators using data structures that mimic existing drug prevention cluster randomized trials and longitudinal studies. Of most interest, cluster size, number of clusters, and degree of non-normality will be varied. Aim 4 applies both Bayesian and existing frequent is to multilevel methods to three existing data sets on drug prevention to compare the performance of the point and interval estimators of the mediation effects from each model. A Monte Carlo comparison of the performance of maximum likelihood and Bayesian approaches as a function of sample size will be conducted using repeated random samples from a large existing data set. The proposal aims to improve statistical methodology in analyzing data from randomized control trials with multiple waves of measurement in the drug prevention areas. The proposed methodology offers new methods when the number of participants is not large and enhances validity of existing statistical methods and the interpretability of the results in drug prevention and health science. My ultimate career goal is to become a faculty member in psychology, education, or the health sciences. I wish to develop and extend quantitative methods that have application in basic psychosocial research and randomized trials on substance abuse prevention. I hope to contribute to the statistical and methodological foundation that will be useful to prevention researchers in understanding the processes through which preventive interventions achieve their effects. PUBLIC HEALTH RELEVANCE: My immediate goals for this fellowship are twofold. First, I will further strengthen my understanding of Bayesian statistics and mediational models. I will continue to do supervised reading under the direction of my committee, notably my co-chairs professors David MacKinnon and Stephen West who are experts in mediation models, longitudinal data analysis, and research design and Professor Roy Levy who is an expert on Bayesian statistics. I have proposed and my committee has approved a comprehensive examination proposal for an extensive paper on Bayesian approaches to mediation which will further strengthen my preparation for the proposed project. I expect to complete my comprehensive examination paper reviewing Bayesian statistical approaches by May, 2009, prior to the beginning of the funding cycle. Second, I will further enhance my training in prevention research, including classes on prevention research methods approaches to prevention and the development of preventive interventions

描述（由申请人提供）：我提出的研究的目标是开发贝叶斯技术，以测试多个测量波的群集随机设计中的中介效应。中介分析研究干预措施通过干预变量实现其效果的过程，这些变量是针对变化的。由于广泛使用的最大似然法依赖于大样本和正常理论来产生有效的结果，贝叶斯技术预计将显示出上级性能在小到中等的样本量，特别是与非正常数据常见的药物滥用预防试验。此外，贝叶斯方法可以结合以前研究的信息，进一步提高其效率。目标1将开发贝叶斯技术来测试集群随机和纵向模型中的中介效应，这些模型可以结合以前实验的信息并处理非正态数据。由此产生的估计可能会导致更可靠的估计和更大的统计能力比目前的方法。目的2发展马尔可夫链蒙特卡罗（MCMC）方法来估计贝叶斯中介模型。此外，在公开的自由软件包（例如，R，WinBUGS）将被编写，使这项工作可供其他研究人员使用。目的3进行模拟研究，比较贝叶斯和最大似然估计的性能，使用模拟现有的药物预防集群随机试验和纵向研究的数据结构。最令人感兴趣的是，聚类大小、聚类数和非正态性程度将有所不同。目的4应用贝叶斯和现有的频繁是多层次方法对现有的三个数据集的药物预防比较点和区间估计的性能从每个模型的中介作用。将使用来自大型现有数据集的重复随机样本，对最大似然法和贝叶斯法作为样本量函数的性能进行蒙特卡罗比较。该提案旨在改进分析药物预防领域多波测量的随机对照试验数据的统计方法。所提出的方法提供了新的方法时，参与者的数量不是很大，并提高了现有的统计方法的有效性和解释性的结果，在药物预防和健康科学。我的最终职业目标是成为心理学，教育学或健康科学的教师。我希望发展和扩展定量方法，应用于基本的心理社会研究和药物滥用预防的随机试验。我希望有助于统计和方法的基础，这将有助于预防研究人员了解预防干预措施实现其效果的过程。公共卫生相关性：我对这个奖学金的直接目标是双重的。首先，我将进一步加强我对贝叶斯统计和中介模型的理解。我将继续在我的委员会的指导下进行监督阅读，特别是我的联合主席教授大卫麦金农和斯蒂芬韦斯特，他们是中介模型，纵向数据分析和研究设计的专家，罗伊利维教授是贝叶斯统计学的专家。我已经提出，我的委员会已经批准了一个全面的审查建议，为一个广泛的文件贝叶斯方法调解，这将进一步加强我的准备拟议的项目。我希望在2009年5月之前完成我的全面考试论文，审查贝叶斯统计方法，在资助周期开始之前。第二，我将进一步加强我在预防研究方面的培训，包括关于预防研究方法、预防办法和制定预防干预措施的课程