DMS-EPSRC Collaborative Research: Advancing Statistical Foundations and Frontiers for and from Emerging Astronomical Data Challenges

DMS-EPSRC 合作研究：为新出现的天文数据挑战推进统计基础和前沿

• 批准号: 2113605
• 负责人: Thomas Chun Man Lee
• 金额: $ 20万
• 依托单位: University of California-Davis
• 依托单位国家: 美国
• 项目类别: Standard Grant
• 财政年份: 2021
• 资助国家: 美国
• 起止时间: 2021-07-01 至 2024-06-30
• 项目状态: 已结题
• 来源: https://www.nsf.gov/awardsearch/showAward?AWD_ID=2113605&HistoricalAwards=false
• 关键词: DMS; EPSRC; Collaborative; Research; Advancing

Statistical theory and methods play a fundamental role in scientific discovery and advancement, including in modern astronomy, where data are collected on increasingly massive scales and with more varieties and complexity. New technology and instrumentation are spawning a diverse array of emerging data types and data analytic challenges, which in turn require and inspire ever more innovative statistical methods and theories. This research is guided by the dual aims of advancing statistical foundations and frontiers, motivated by astronomical problems and providing principled data analytic solutions to challenges in astronomy. The CHASC (California-Harvard Astrostatistics Collaboration) International Center has an extensive track record in accomplishing both tasks. This research leverages CHASC’s track record to make progress in several new projects. Fitting sophisticated astrophysical models to complex data that were collected with high-tech instruments, for example, often involves a sequence of statistical analyses. Several projects center on developing new statistical methods that properly account for errors and carry uncertainty forward within such sequences of analyses. Additional work will focus on developing theoretical properties of novel statistical estimation procedures to address data-analytic challenges associated with solar flares and X-ray observations. Other projects involve fast and automatic detection of astronomical objects such as galaxies from 2D or even 4D data. The PIs will develop statistical theory and methods in the context of these projects, building statistical foundations and pushing the frontiers of statistics forward for broad impact that will extend well beyond astrostatistics. The PIs plan to offer effective methods and algorithms for tackling emerging challenges in astronomy, with the aspiration of promoting such principled data-analytic methods among researchers in astronomy. Its provision of free software via the CHASC GitHub Software Library will enable the distribution and impact of the proposed methods and algorithms. The projects reflect three broad themes: (1) Exploring fundamental statistical theory with immediate impact in astronomy, including a general approach for obtaining confidence regions by leveraging the pivot-property of maximal product spacing, which is then applied to assess the power law of solar flares, and a statistically principled correction to the use of the popular C-stat in astrophysics; (2) Assessing the misspecification of models and prior distributions in multi-stage statistical analyses, and post processing posterior draws to correct for defects in prior modeling when redoing a Bayesian analysis is impractical; and (3) Identifying breakpoints in complex models, which includes a fast algorithm for identifying astronomical boundaries and identifying breakpoints in joint spatial, spectral, temporal models. Theme 1 is more theory driven, while Themes 2 and 3 are more methods and computation driven. Together they form a rich suite of case studies for developing statistical methods for astronomical problems, ranging from new theoretical foundations to innovative modeling strategies and to efficient computational techniques. Consequently, the research will impact both the fields of statistics and astronomy: spurring more interest and new problems for statisticians and resolving long standing problems in astronomy.This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.

统计理论和方法在科学发现和进步中发挥着重要作用，包括在现代天文学中，收集的数据规模越来越大，种类越来越多，越来越复杂。新的技术和工具正在催生各种各样的新兴数据类型和数据分析挑战，这反过来又需要并激发更多创新的统计方法和理论。这项研究的双重目标是推进统计基础和前沿，由天文学问题和提供原则性的数据分析解决方案，以应对天文学的挑战。CHASC（加州-哈佛天体统计合作）国际中心在完成这两项任务方面有着广泛的记录。这项研究利用CHASC的跟踪记录在几个新项目中取得进展。例如，将复杂的天体物理模型与高科技仪器收集的复杂数据相匹配，通常涉及一系列统计分析。有几个项目集中于开发新的统计方法，这些方法可以适当地解释错误，并在这样的分析序列中向前推进不确定性。其他工作将侧重于开发新的统计估计程序的理论特性，以解决与太阳耀斑和X射线观测相关的数据分析挑战。其他项目涉及从2D甚至4D数据中快速自动检测天文物体，如星系。研究所将在这些项目的背景下发展统计理论和方法，建立统计基础，推动统计前沿，产生远远超出天体统计的广泛影响。PI计划提供有效的方法和算法来应对天文学中的新挑战，并希望在天文学研究人员中推广这种原则性的数据分析方法。它通过CHASC GitHub软件库提供免费软件，将使所提出的方法和算法的分发和影响成为可能。这些项目反映了三个广泛的主题：（1）探索对天文学有直接影响的基本统计理论，包括利用最大乘积间距的非线性特性获得置信区的一般方法，然后将其用于评估太阳耀斑的幂律，以及对天体物理学中流行的C统计的使用进行统计原则性校正;（2）在多阶段统计分析中评估模型和先验分布的错误指定，并且当重做贝叶斯分析不切实际时，后处理后验绘制以校正先验建模中的缺陷;以及（3）识别复杂模型中的断点，其包括用于识别天文边界和识别联合空间，光谱，时间模型主题1更多的是理论驱动，而主题2和3更多的是方法和计算驱动。他们一起形成了一套丰富的案例研究，为天文学问题开发统计方法，从新的理论基础到创新的建模策略，再到高效的计算技术。 因此，该研究将同时影响统计学和天文学领域：激发统计学家的更多兴趣和新问题，并解决天文学中长期存在的问题。该奖项反映了NSF的法定使命，并通过使用基金会的知识价值和更广泛的影响审查标准进行评估，被认为值得支持。

项目成果

Statistical Consistency for Change Point Detection and Community Estimation in Time-Evolving Dynamic Networks

• DOI: 10.1109/tsipn.2022.3156434
• 发表时间: 2022
• 期刊: IEEE Transactions on Signal and Information Processing over Networks
• 影响因子: 3.2
• 作者: Cong Xu;Thomas C.M. Lee

Uncertainty Quantification in Graphon Estimation Using Generalized Fiducial Inference

使用广义基准推理的图估计中的不确定性量化

• DOI: 10.1109/tsipn.2022.3188458
• 发表时间: 2022
• 期刊: IEEE Transactions on Signal and Information Processing over Networks
• 影响因子: 3.2
• 作者: Su, Yi;Hannig, Jan;Lee, Thomas C.
• 通讯作者: Lee, Thomas C.

Uncertainty Quantification for Sparse Estimation of Spectral Lines

谱线稀疏估计的不确定性量化

• DOI: 10.1109/tsp.2023.3235662
• 发表时间: 2022
• 期刊: IEEE Transactions on Signal Processing
• 影响因子: 5.4
• 作者: Han, Yi;Lee, Thomas C.
• 通讯作者: Lee, Thomas C.

Identifying Diffuse Spatial Structures in High-energy Photon Lists

识别高能光子列表中的漫射空间结构

• DOI: 10.3847/1538-3881/aca478
• 发表时间: 2023
• 期刊: The Astronomical Journal
• 作者: Fan, Minjie;Wang, Jue;Kashyap, Vinay L.;Lee, Thomas C. M.;van Dyk, David A.;Zezas, Andreas
• 通讯作者: Zezas, Andreas

Change Point Detection and Node Clustering for Time Series of Graphs

图时间序列的变化点检测和节点聚类

• DOI: 10.1109/tsp.2022.3183359
• 发表时间: 2022
• 期刊: IEEE Transactions on Signal Processing
• 影响因子: 5.4
• 作者: Xu, Cong;Lee, Thomas C.
• 通讯作者: Lee, Thomas C.