Statistical Inference and Modeling for Complex Data

复杂数据的统计推断和建模

• 批准号: 0072840
• 负责人: Jiayang Sun
• 金额: $ 10万
• 依托单位: Case Western Reserve University
• 依托单位国家: 美国
• 项目类别: Standard Grant
• 财政年份: 2000
• 资助国家: 美国
• 起止时间: 2000-08-15 至 2004-01-31
• 项目状态: 已结题
• 来源: https://www.nsf.gov/awardsearch/showAward?AWD_ID=0072840&HistoricalAwards=false
• 关键词: Statistical; Inference; Modeling; Complex; Data

This project focuses on three research areas related to complex data.The first research area is on high dimensional graphics, data analysisand related topics. The second research area is on mixture models andbump hunting problems. The third area is on models and inferentialprocedures for data with biases. Several variants of likelihood,semi-parametric and iterative procedures are sought for modeling andmaking inferences about the data.We live in the information age. Information often hides in data setswith different appearances. Mining and modeling (massive) data setsinclude the discovery of patterns, subcomponents, bumps, or specialevents, and development of models that explain the data and predictthe future. This research is aimed at providing solutions to someinteresting problems arising from the data which are either highdimension, or complicated, or with a sampling bias. Theories,methods, and efficient computing algorithms are explored forextracting useful information from the complex data. Applicationareas include astronomy, neurosciences, quality control, and some(other) observational studies.

本计画主要针对复杂资料之相关研究领域，第一个研究领域为高维图形、资料分析及相关议题。 第二个研究领域是混合模型和颠簸狩猎问题。 第三个领域是关于模型和带有偏差的数据的推理程序。 几个变量的可能性，半参数和迭代程序的建模和推理的数据。我们生活在信息时代。 信息往往隐藏在不同外观的数据集中。 挖掘和建模（海量）数据集包括发现模式、子组件、碰撞或特殊事件，以及开发解释数据和预测未来的模型。 本研究的目的是为解决高维、复杂、有抽样偏差的数据所引起的一些有趣的问题提供解决方案。 理论，方法， 高效计算 算法 探索从复杂的数据中提取有用的信息。 应用领域包括天文学，神经科学，质量控制和一些（其他）观测研究。