Symposium on Mathematical Statistics and Applications: From Time Series and Stochastics, to Semi- and Non-Parametrics, to High-Dimensional Models

该奖项支持参加于2018年9月14-16日在密歇根州立大学举行的“数理统计与应用研讨会：从时间序列和随机，到半参数和非参数，再到高维模型”的会议。这次会议的目的是让数理统计领域的研究生和早期研究人员受益。随着21世纪的数据革命将美国带入商业和政府中基于数据的决策和预测分析的新领域，受过数学培训并注重分析的统计学家的作用仍然比以往任何时候都重要，他们开发和分析合理、创新的方式来理解数据科学的算法、方法和程序。这次会议将汇集来自国内外的知名和有抱负的研究人员，探索数理统计的前沿和一系列新兴的应用。这次会议为数十名研究生和职业生涯初期的研究人员提供的机会将是非常有价值的，可能会改变一些人的职业生涯。在几个领域，特别是在半参数和非参数统计领域，正在开展基础工作，以进一步分析和使用各种形式的数据。这样的数学发展需要跨越统计科学的广泛领域，包括高维统计推断、非参数时间序列、函数数据分析、经验过程和机器学习。会议将向与会者介绍这些主题以及其他方向，包括自适应和有效估计的渐近理论、计量经济学中的时间序列和生存分析。由于讨论日程安排中有充足的时间，会议将使与会者有机会进行富有成效的跨群体合作。这次会议的一个特别有影响力的方面是，当地非统计人员作了几次演讲，他们在通常没有得到统计人员服务的领域工作，但有非常重要的数据分析需求，他们都对与全国各地的统计人员互动感兴趣，而这些统计人员通常没有机会见面。涵盖的应用领域将包括定量金融、保险数学和风险分析、核物理、农业经济学、神经成像、土壤科学、医院管理、水文学等。会议网站是https://stt.msu.edu/MSUStatSymposium2018/.This奖，反映了国家科学基金会的法定使命，并通过使用基金会的智力优势和更广泛的影响审查标准进行评估，被认为值得支持。

This award supports participation in the conference "Symposium on Mathematical Statistics and Applications: From Time Series and Stochastics, to Semi- and Non-Parametrics, to High-Dimensional Models" held September 14 - 16, 2018 at Michigan State University. This conference is designed to benefit graduate students and early-career researchers in mathematical statistics. As the data revolution of the 21st century takes the US to new frontiers in data-based decision-making and predictive analytics in business and government, the role of the mathematically-trained and analytically-critical statistician remains more important than ever, to develop and analyze sound, innovative ways of understanding data science algorithms, methodologies, and procedures. The conference will bring together established and aspiring researchers from around the country and abroad to explore frontiers of mathematical statistics and a host of emerging applications. The exposure and networking opportunities provided by the conference to the dozens of graduate students and early-career researchers will be of great value, and potentially career-changing for some.Foundational work to further develop the analysis and use of data in all its forms is underway in several areas, particularly in semi- and non-parametric statistics. Such mathematical development is needed across a wide spectrum of the statistical sciences, including high-dimensional statistical inference, non-parametric time series, functional data analysis, empirical processes, and machine learning. The conference will expose the participants to these topics as well as other directions, including asymptotic theory of adaptive and efficient estimation, time series in econometrics, and survival analysis. With ample time built into the schedule for discussions, the conference will give participants opportunities to engage in fruitful cross-group collaborations. A particularly impactful aspect of the conference is the inclusion of several presentations by local non-statisticians who work in areas typically underserved by statisticians but with very significant data analysis needs, all interested in interacting with statisticians from across the country with whom there would ordinarily be no chance to meet. Application areas covered will include quantitative finance, insurance mathematics and risk analytics, nuclear physics, agricultural economics, neuro-imaging, soil science, hospital management, hydrology, and others. The conference website is https://stt.msu.edu/MSUStatSymposium2018/.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.