Biography:  Songxi Chen (Chinese: 陈松蹊), Chair Professor of Peking University; Member of the Chinese Academy of Sciences; Fellow of the Institute of Mathematical Statistics, the American Statistical Association and the American Association for the Advancement of Science; Elected Member of the International Statistical Institute; Scientific Secretary of the Bernoulli Society.

Title: Experimental and Statistical Analysis of Data in the Atmospheric Environment

Abstract: The talk presents research effort designed to extract underlying emission information from air pollution monitoring network data, and part of our eight-year effort on the analysis and tracking of air pollution situation in northern China. The talk gives mathematical and statistical formulation as well as data experiments constructed to gain underlying emission information from meteorologically confounded air quality data. This includes methods to remove meteorological confounding, indices to measure man-made vs natural meteorological effects on the air pollution, and data experiments to measure local emissions. The presentation will also reflect on the current paradigm of statistical research and explore the paradigm of atmospheric environment research useful for economics and statistical research.

摘要：报告从基于大气污染监测网络数据提取污染排放信息为主旨，介绍团队八年来分析、追踪北方地区大气污染的变化的实证研究， 给出大气污染评估的数学思路和统计学实现方法。包括去除大气污染观测数据中的气象干扰方法，度量大气污染变化的"人努力-天帮忙"指数，和度量供暖效应、重大活动和本地排放对空气质量影响的数据实验。报告也将反思目前统计学的研究范式，探讨大气环境的研究范式对经济学和统计学研究的可借鉴之处。