权益分类	功能权益	普通用户	{{item.name}}会员
{{category.name}}	{{benefitItem.name}}

Non-Parametric Bayesian Methods for Causal Inference

用于因果推理的非参数贝叶斯方法

基本信息

批准号：
8925116
负责人：
JASON A ROY
金额：
$ 34.4万
依托单位：
UNIVERSITY OF PENNSYLVANIA
依托单位国家：
美国
项目类别：
财政年份：
2014
资助国家：
美国
起止时间：
2014-09-10 至 2018-06-30
项目状态：
已结题

来源：
https://reporter.nih.gov/project-details/8925116
关键词：
Accounting Adoption Affect Aging Algorithms Anti-Retroviral Agents Area Bayesian Method Behavior Therapy Behavioral Research Behavioral trial Categories Chronic Hepatitis C Clinical Clinical Research Code Cohort Studies Comparative Study Complex Computer software Data Development Documentation Dropout Evaluation Failure Future Goals HIV Health Hepatic Hepatitis C virus Internet Intervention Intervention Trial Journals Lead Literature Mediating Mediation Mediator of activation protein Methods Modeling Nature Observational Study Outcome Patients Performance Pharmaceutical Preparations Population Publishing Reproducibility Research Research Personnel Risk Safety Sampling Specific qualifier value Statistical Methods Survival Analysis Time Uncertainty Veterans Weight maintenance regimen Work base causal model clinically relevant comparative effectiveness falls flexibility interest method development non-compliance novel novel strategies open source randomized trial research study simulation smoking cessation sound treatment strategy web site

项目摘要

DESCRIPTION (provided by applicant): The overarching goal of this project is to develop Bayesian non-parametric (BNP) methods for estimating causal effects from complex data. We focus on two broad areas: survival analysis with time-varying treatments and mediation. For survival outcomes, we develop BNP methods for estimating causal parameters from structural nested failure time models, both for discrete and continuous-time problems. Likelihood-based methods have generally not been implemented for these models, because it would require many parametric modeling assumptions. Our BNP approach should provide greater flexibility than parametric models, while maintaining computational advantages. We will develop these methods for a wide array of scenarios (e.g., multinomial or continuous-valued treatment, known or unknown censoring times) and develop sensitivity analysis methods and informative priors related to untestable assumptions. For causal mediation analysis, we will extend our previous work in a variety of ways. Most importantly, we will weaken identifying assumptions with the inclusion of covariates in the models. In addition, we will generalize to a wider variety of outcomes and types of mediation (e.g. longitudinal or multiple mediators). We will also develop methods for handling non-ignorable dropout in settings with mediation. Our methods have broad applications, and we will utilize them to draw novel clinical inference from several behavioral intervention trials and from a study on the hepatic safety of classes of antiretroviral medications

描述（由申请人提供）：该项目的总体目标是开发贝叶斯非参数（BNP）方法，用于从复杂数据中估计因果效应。我们专注于两个广泛的领域：随时间变化的治疗和调解的生存分析。对于生存结果，我们开发BNP方法估计因果参数的结构嵌套故障时间模型，离散和连续时间的问题。基于似然性的方法通常不适用于这些模型，因为它需要许多参数建模假设。我们的BNP方法应该提供比参数模型更大的灵活性，同时保持计算优势。我们将为各种场景开发这些方法（例如，多项式或连续值处理，已知或未知的删失时间），并开发敏感性分析方法和与不可检验的假设相关的信息先验。对于因果中介分析，我们将以各种方式扩展我们以前的工作。最重要的是，我们将通过在模型中纳入协变量来削弱识别假设。此外，我们将归纳到更广泛的结果和调解类型（例如纵向或多个调解人）。我们还将开发在具有中介的设置中处理不可解释的辍学的方法。我们的方法具有广泛的应用，我们将利用它们从几个行为干预试验和一项关于抗逆转录病毒药物的肝脏安全性的研究中得出新的临床推断