Mathematical Sciences: Studies in Nonparametric and Semiparametric Statistics

数学科学：非参数和半参数统计研究

• 批准号: 8901603
• 负责人: Kjell Doksum
• 金额: $ 6.3万
• 依托单位: University of California-Berkeley
• 依托单位国家: 美国
• 项目类别: Continuing Grant
• 财政年份: 1989
• 资助国家: 美国
• 起止时间: 1989-06-15 至 1991-05-31
• 项目状态: 已结题
• 来源: https://www.nsf.gov/awardsearch/showAward?AWD_ID=8901603&HistoricalAwards=false
• 关键词: Mathematical; Sciences; Studies; Nonparametric; Semiparametric

The research involves the study of semiparametric and nonparametric statistical models in the presence of censored data. Graphical model diagnostic methods will be developed to compare the hazard experience of two groups in two-sample experiments and for goodness-of-fit tests of the Cox regression model. Confidence band procedures will be developed for nonparametric regression and data-based rules will be given in bandwidth selection problems for hazard function estimation. The relationships between the response variable and predictor variables will be studied through semparametric rate functions and transformation models. Procedures will also be considered for testing independence of paired survival times.

该研究涉及到半参数和非参数统计模型在审查数据存在的研究。将开发图形模型诊断方法，以比较两组在双样本实验中的危险经验，并进行Cox回归模型的拟合优度检验。我们将开发用于非参数回归的置信带程序，并给出用于风险函数估计的带宽选择问题的基于数据的规则。通过半参数速率函数和变换模型研究响应变量和预测变量之间的关系。程序也将考虑测试成对生存时间的独立性。