Robust Missing Data Methods for Categorical Regression

用于分类回归的稳健缺失数据方法

• 批准号: 6549395
• 负责人: Steven S Henley
• 金额: $ 10万
• 依托单位: MARTINGALE RESEARCH CORPORATION
• 依托单位国家: 美国
• 财政年份: 2002
• 资助国家: 美国
• 起止时间: 2002-09-25 至 2003-08-31
• 项目状态: 已结题
• 来源: https://reporter.nih.gov/project-details/6549395
• 关键词: alcoholic beverages; alcoholism /alcohol abuse; alcoholism /alcohol abuse information system; analytical method; computer program /software; data collection methodology /evaluation; human data; mathematical model; mathematics; method development; model design /development; statistics /biometry; substance abuse epidemiology

DESCRIPTION (provided by applicant): Improved methods for obtaining robust statistical inferences from categorical regression models in the presence of missing data and model misspecification would be an invaluable tool to the epidemiological and health care research communities. Presently epidemiological models are typically designed to identify patterns of alcohol-related symptoms, define criteria of alcohol use disorders, and evaluate policies regulating use and distribution of alcoholic beverages. Such models frequently rely on datasets that contain incomplete-data. While commercially available statistical software provides some automated missing value procedures (e.g., data imputation, Expectation-Maximization), further theoretical and empirical research is required to develop more robust statistical methods. Martingale Research will develop Robust Expectation-Maximization (REM) algorithms that combine recent advances in stochastic estimation, asymptotic statistics, and maximum likelihood recoding to achieve the following objectives: 1) design and implement REM algorithms that are suited to categorical regression modeling for epidemiological problems, 2) theoretically and empirically investigate REM algorithm performance in the presence of missing data and model misspecification and 3) extend the REM algorithms to optimally estimate preprocessing transformations in the presence of missing data and model misspecification. These results will demonstrate the essential technical feasibility required for further Phase II investigations and provide the foundation for developing commercially available software.

描述（由申请人提供）：在存在缺失数据和模型错误指定的情况下，从分类回归模型中获得稳健统计推断的改进方法将是流行病学和医疗保健研究界的宝贵工具。目前流行病学模型通常被设计用于识别酒精相关症状的模式，定义酒精使用障碍的标准，并评估规范酒精饮料使用和销售的政策。这些模型通常依赖于包含不完整数据的数据集。虽然商业上可获得的统计软件提供了一些自动缺失值程序（例如，数据填补，期望最大化），需要进一步的理论和实证研究，以开发更强大的统计方法。 Martingale Research将开发鲁棒期望最大化（REM）算法，该算法联合收割机了随机估计，渐近统计和最大似然重编码的最新进展，以实现以下目标：1）设计并实现适合于流行病学问题的分类回归建模的REM算法，2）在存在缺失数据和模型错误指定的情况下，从理论上和经验上研究REM算法性能，以及3）扩展REM算法，以最佳估计预处理转换的缺失数据和模型误指定的存在。这些结果将证明进一步的第二阶段调查所需的基本技术可行性，并为开发商业软件提供基础。

项目成果

Measuring use and cost of care for patients with mood disorders: the utilization and cost inventory.

衡量情绪障碍患者的护理使用和成本：利用率和成本清单。

• DOI: 10.1097/mlr.0b013e31818457b8
• 发表时间: 2009
• 期刊: Medical care
• 影响因子: 3
• 作者: Kashner,TMichael;Stensland,MichaelD;Lind,Lisa;Wicker,Annie;Rush,AJohn;Golden,RichardM;Henley,StevenS
• 通讯作者: Henley,StevenS