Robust Classification Methods for Categorical Regression

分类回归的稳健分类方法

• 批准号: 7686932
• 负责人: Steven S Henley
• 金额: $ 95.79万
• 依托单位: MARTINGALE RESEARCH CORPORATION
• 依托单位国家: 美国
• 财政年份: 2003
• 资助国家: 美国
• 起止时间: 2003-06-04 至 2011-08-31
• 项目状态: 已结题
• 来源: https://reporter.nih.gov/project-details/7686932
• 关键词: Accident and Emergency department; Achievement; Address; Administrator; Agreement; Alcohols; Algorithms; Area; Bayesian Method; Behavior; Bipolar Depression; Classification; Clinical; Clinical Investigator; Collaborations; Communicable Diseases; Communities; Complex; Computer software; Confidence Intervals; Data; Data Analyses; Data Set; Decision Analysis; Decision Making; Development; Diagnosis; Disease; Empirical Research; Engineering; Epidemiologic Studies; Epidemiologist; Epidemiology; Evaluation; Event; Foundations; Goals; Guidelines; Health; Health Services Research; Healthcare; Industry; Information Resources Management; Injury; Jordan; Journals; Knowledge; Literature; Logistic Regressions; Logistics; Malignant Neoplasms; Malignant neoplasm of prostate; Medical; Mental Health; Mental disorders; Methodology; Methods; Modeling; Nature; Outcome; Output; Pathology; Pattern; Peer Review; Performance; Phase; Phase I Clinical Trials; Phase II Clinical Trials; Policies; Preparation; Probability; Process; Publishing; Quality of Care; Relative (related person); Research; Research Personnel; Risk; Robin bird; Sampling; Schizophrenia; Screening procedure; Sensitivity and Specificity; Simulate; Software Tools; Specific qualifier value; Specificity; Statistical Methods; Surveys; Symptoms; Technology; Testing; Trauma; anticancer research; base; commercialization; computerized data processing; density; design; graphical user interface; improved; innovation; prototype; public health research; simulation; software development; theories; tool; user friendly software

DESCRIPTION (provided by applicant): Improving statistical methods to provide better classification performance and new analytical capabilities for categorical regression would be invaluable to the medical and health care research communities. Categorical regression models (e.g., binary logistic, multinomial logistic) are used extensively to identify patterns of alcohol-related symptoms, screen for disorders, and assess policies. In addition, such models are used extensively in other areas of research such as mental illness, cancer, traumatic injuries, and AIDS-related pathologies. However, many such models are developed with inadequate support to fully analyze and exploit the intrinsically probabilistic nature of their results. This is of critical importance as health researchers, clinicians, and administrators are often faced with classification decisions using categorical regression models to identify unacceptable risks, adequate outcomes, and acceptable guidelines for screening, diagnoses, treatment, and quality of care. Commercially available statistical software does not offer sophisticated methods for robust estimation of posterior probabilities in the presence of model misspecification, missing covariates, and nonignorable missing data generating processes. Such robust missing data handling methods provide natural mechanisms for dealing with verification bias and modeling correlated, longitudinal, or survey data with complex sampling designs. Moreover, commercially available statistical software does not provide automated methods for using estimated posterior probabilities to make optimal classification decisions with respect to different optimality criteria. In particular, automated features such as optimizing multiple decision criteria (allocation rules) that trade off specificity against sensitivity, decision threshold confidence intervals, statistical tests for evaluating correct specification of posterior probabilities, statistical tests for comparing competing classifier thresholds, and methods for multi-outcome classification and inference are not readily available. Phase II research will extend Phase I findings for binary logistic regression to develop and implement automated robust classification methods for multinomial logistic regression modeling, which also applies to the larger class of nonlinear categorical regression models that output posterior probabilities. The Phase II software prototype will provide: 1) new user-selectable robust decision threshold estimators, 2) robust confidence intervals on decision threshold estimators, 3) new classifier threshold comparison tests, 4) new outcome probability specification tests, 5) efficient missing data handling methods in the presence of nonignorable nonresponse data, and 6) second-order analytic and simulation-based Bayesian methods for improved small sample and rare event outcome probability estimation. These new methodologies will be integrated into a prototype user-friendly software package, evaluated with extensive simulation studies, and then applied to real world classification problems encountered in: alcohol, mental illness (depression, bipolar, schizophrenia), cancer (prostate), trauma (emergency room), and infectious disease (AIDS) through collaborations with domain experts in those respective fields. In summary, Phase II research will establish the essential technical foundation for Phase III commercialization with the objective of providing a suite of new classification analysis methods as an advanced statistical tool that improves epidemiologic, clinical, and public health research.

描述（由申请人提供）：改进统计方法以提供更好的分类性能和新的分类回归分析能力，对医学和卫生保健研究界将是无价的。分类回归模型（例如，二元逻辑、多项逻辑）被广泛用于识别酒精相关症状的模式、筛查障碍和评估政策。此外，这些模型还广泛应用于其他研究领域，如精神疾病、癌症、创伤性损伤和艾滋病相关病理。然而，许多这样的模型是在没有充分支持的情况下开发的，无法充分分析和利用其结果的内在概率性质。这是至关重要的，因为卫生研究人员、临床医生和管理人员经常面临使用分类回归模型来确定不可接受的风险、适当的结果和可接受的筛查、诊断、治疗和护理质量指南的分类决策。商业上可用的统计软件不提供复杂的方法，在模型错误规范，缺失协变量和不可忽略的缺失数据生成过程中，对后验概率进行稳健估计。这种健壮的缺失数据处理方法为处理验证偏差和建模相关的、纵向的或具有复杂抽样设计的调查数据提供了自然机制。此外，商业上可用的统计软件并没有提供自动化的方法来使用估计的后验概率来根据不同的最优性标准做出最优分类决策。特别是，自动化特征，如优化多个决策标准（分配规则），权衡特异性与敏感性，决策阈值置信区间，评估后验概率正确规范的统计测试，比较竞争分类器阈值的统计测试，以及多结果分类和推理的方法，都不容易获得。第二阶段的研究将扩展第一阶段关于二元逻辑回归的研究成果，以开发和实现多项逻辑回归模型的自动鲁棒分类方法，这也适用于输出后验概率的更大类非线性分类回归模型。第二阶段的软件原型将提供：1)新的用户可选择的鲁棒决策阈值估计器，2)决策阈值估计器的鲁棒置信区间，3)新的分类器阈值比较测试，4)新的结果概率规范测试，5)在存在不可忽略的非响应数据时有效的缺失数据处理方法，以及6)二阶分析和基于模拟的贝叶斯方法，用于改进小样本和罕见事件结果概率估计。这些新方法将被整合到一个用户友好的原型软件包中，通过广泛的模拟研究进行评估，然后通过与这些领域的领域专家合作，将其应用于现实世界中遇到的分类问题：酒精、精神疾病（抑郁症、双相情感障碍、精神分裂症）、癌症（前列腺癌）、创伤（急诊室）和传染病（艾滋病）。总之，二期研究将为三期商业化奠定必要的技术基础，目的是提供一套新的分类分析方法，作为一种先进的统计工具，改善流行病学、临床和公共卫生研究。