Advances in Experimental Particle Physics through Statistical Methodology and Data Analysis

经典统计方法是在20世纪的大部分时间里发展起来的，用来分析小规模到中等规模的数据集。功能大大增加的计算设备的无处不在导致了海量数据集的收集和存储。在许多领域，如电信和粒子物理，数据集本质上是无限的：随着连接的建立、金融交易的进行以及粒子随着时间的推移而相互作用，新数据不断涌现。此外，数据可能涉及许多高度相关的变量；例如，除了交易金额，交易数据可能包括关于购买者和供应商的人口统计变量，以及物理实验中粒子的数据涉及速度、动量和质量等特征，所有这些都会影响其他粒子的特征。如此庞大的数据集需要开发新的统计方法。需要这种方法的一个重要领域是高能粒子物理。提出的研究的目标是开发用于查看和识别海量数据中的模式的统计方法和算法，具体目标是将这些方法应用于来自粒子物理实验的数据，并由此导致：(1)通过实验数据库的新用途在粒子物理理论方面的进展，(2)用于分析海量数据集的新的统计方法，(3)为计划在物理科学领域进行跨学科研究的学生提供增强的教育经验。PI将加强她在粒子物理方面的本科教育，并在美国国家标准局(现为NIST)的统计工程部和惠普公司(现为安捷伦)的微波测试产品和研究部门，在斯坦福大学斯坦福大学直线加速器中心(SLAC)巴布尔合作中心和伯克利(劳伦斯·伯克利实验室和物理系)的首席研究员罗伯特·G·雅各布森教授的指导下，加强她在粒子物理方面的本科教育和工作经验。回到丹佛后，她将向统计学和物理学的学生介绍高能物理中的问题和分析海量数据集的统计方法，并将这一经验应用于NIST-Boulder时间和频率分部的一级和二级频率标准(铯喷泉；五个氢脉泽，四个商用铯标准)校准模型中的不确定度评估问题。在物理学上，这项研究应该会导致与B介子(SLAC)相关的特定粒子衰变反应相关的不确定性的更高精度和量化。这些知识将有助于国际空间局与NIST-Boulder的物理学家就铯喷泉频率标准的不确定度进行合作。在统计学方面，这项研究应该会导致分析来自这类实验的大量数据的新方法。更广泛的影响：国际和平研究所预计，这一合作不仅将有助于物理学家对涉及B介子的衰变反应的了解，还将引领物理学家使用粒子物理实验数据的方法，促进对标准模型的了解，导致分析海量数据集/流的新方法，并增强学生在计算统计方面的教育经验。该IGMS项目由重大计划多学科活动办公室(OMA)和数学科学处(DMS)共同支持。

Classical statistical methods were developed during much of the 20th century to analyze small-to-moderatesized data sets. The ubiquity of computing devices with vastly increased capabilities has led to the collection and storage of massive data sets. In many fields, such as telecommunications and particle physics, the data sets are essentially infinite: new data arise continuously as connections are established, financial transactions are conducted, and particles interact over time. Moreover, the data are likely to involve manyhighly correlated variables; e.g., data on a transaction may include demographic variables on both the purchaser and the supplier, in addition to the amount of the transaction, and data on particles in a physics experiment involve characteristics such as velocity, momentum, and mass, all of which can affect the characteristics on other particles.Such massive data sets require the development of new statistical methods. One important area where suchmethods are needed is high-energy particle physics. The goal for the proposed research is to develop statistical methods and algorithms for viewing, and identifying patterns in, massive amounts of data,with the specific goal of applying these methods to data from particle physics experiments, and hencelead to: (1) advances in the theory of particle physics through novel uses of experimental databases, (2) new statistical methods for analyzing massive data sets, and (3) enhanced educational experiences for students who plan to pursue interdisciplinary research in the physical sciences.The PI will enhance her undergraduate education in particle physics and work experience in the Statistical Engineering Division at the National Bureau of Standards (now NIST) and in the microwave test products and research division at Hewlett Packard Company (now Agilent), under the supervision of Professor Robert G. Jacobsen, a lead researcher at Stanford Stanford Linear Accelerator Center (SLAC) BaBar collaborationand in Berkeley (Lawrence Berkeley Laboratory and Department of Physics). Upon returning to Denver, she will introduce students in both statistics and physics to problems in high-energy physics and statistical methodology for the analysis of massive data sets, and apply this experience to the problem of assessment of uncertainty in the calibration models for the primary and secondary frequency standards (cesium fountain; five hydrogen masers, four commercial cesium standards) at NIST-Boulder's Time and Frequency Division.Intellectual Merit:Specific outcomes of the proposed research include advances in both physics and statistics. In physics,this research should lead to greater precision and quantification of uncertainties associated with specific particle decay reactions associated with the B meson (SLAC). This knowledge will be useful in thePI's collaborations with physicists at NIST-Boulder on the uncertainty in the cesium fountain frequency standard. In statistics, this research should lead to new methods for analyzing massive data from such experiments. Broader impacts:The PI expects that this collaboration will benefit not only physicists' understanding of decay reactions involving the B meson but also will lead the way for physicists' use of data from particle physics experiments, advance the state of knowledge about the Standard Model, lead to new methods of analyzing massive data sets/streams, and enhance the educational experience for students in computational statistics. This IGMS project is jointly supported by the MPS Office of Multidisciplinary Activities (OMA) and the Division of Mathematical Sciences (DMS).