Mathematical Sciences: Multivariate and Censored Data Analysis Methods for Astronomy

数学科学：天文学的多元和审查数据分析方法

• 批准号: 9208066
• 负责人: Michael Akritas
• 金额: $ 16万
• 依托单位: Pennsylvania State Univ University Park
• 依托单位国家: 美国
• 项目类别: Continuing Grant
• 财政年份: 1992
• 资助国家: 美国
• 起止时间: 1992-08-15 至 1996-07-31
• 项目状态: 已结题
• 来源: https://www.nsf.gov/awardsearch/showAward?AWD_ID=9208066&HistoricalAwards=false
• 关键词: Mathematical; Sciences; Multivariate; Censored; Data

This project continues collaborative research between astronomers and statisticians with the objective of developing specifically useful statistical procedures for the analysis of multivariate data sets that arise in observational astronomy. It addresses the statistical problems of estimation of a bivariate distribution function under random censoring, censored regression, testing for multivariate normality under random censoring and a new analysis for statistical independence. Statistical analysis can elucidate astrophysical processes which are detectable only from data obtained from a sample of celestial objects. Using modern statistical theory and methodology, some of the obscuring distortions which are inevitable in such data can be "removed" to make the most physically meaningful relations more apparent. For example, the relations between luminosity, size, rotation and metallicity in elliptical galaxies can be examined and described. This collaborative research focuses on the development of new methods that carry the potential for major advances in both statistics and astronomy.

该项目继续进行天文学家和统计学家之间的合作研究，目的是为分析观测天文学中出现的多变量数据集制定特别有用的统计程序。它讨论了随机截尾下的二元分布函数的估计、截尾回归、随机截尾下的多元正态检验以及一种新的统计独立性分析的统计问题。统计分析可以阐明只有从天体样本获得的数据才能检测到的天体物理过程。利用现代统计理论和方法，这些数据中不可避免的一些模糊的扭曲可以被“去除”，以使最具物理意义的关系更加明显。例如，可以研究和描述椭圆星系中光度、大小、自转和金属丰度之间的关系。这项合作研究的重点是开发新的方法，这些方法具有在统计学和天文学方面取得重大进展的潜力。