Files in This Item:
File Format
b1552003.mp4Streaming VideoView/Open
Title: Statistical Paradises and Paradoxes in Big Data
Originating Office: IAS
Speaker: Meng, Xiao-Li
Issue Date: 18-Jan-2016
Event Date: 18-Jan-2016
Group/Series/Folder: Record Group 8.15 - Institute for Advanced Study
Series 3 - Audio-visual Materials
Location: 8.15:3 EF
Notes: IAS distinguished lecture.
Title from opening screen.
Abstract: Statisticians are increasingly posed with thought- provoking and often paradoxical questions, challenging our qualifications for entering the statistical paradises created by Big Data. In this talk, the speaker will address two questions: (1) 'Which one should I trust: a 1% survey with 60% response rate or a self-reported administrative dataset covering 80% of the population?' and (2) 'Personalized treatments that sounds heavenly, but where on earth did they find the right guinea pig for me?' The proper responses are respectively (1) 'It depends!', because we need data-quality indexes, not merely quantitative sizes, to determine; and (2) 'They didn't!', but the question has led to a multi-resolution framework for studying statistical evidence for predicting individual outcomes. These questions highlight the need, as getting deeper into this era of Big Data, to reaffirm some time-honored statistical themes (e.g., bias-variance trade-off), and to remodel some others (e.g., approximating individuals from proxy populations verses inferring populations from samples).
Prof Xiao-Li Meng received his PhD in Statistics from Harvard University in 1991. He joined the University of Chicago as an Assistant Professor and eventually became a full Professor before he left Chicago in 2005. In 2001, Prof Meng returned to Harvard as a Professor, where he was also the Chair of the Department of Statistics in 2004-2012. He is currently the Dean of the Graduate School of Arts and Sciences and Whipple V. N. Jones Professor of Statistics in Harvard University.
Prof Meng’s research focuses on a wide variety of topics, including statistical inference with partially observed data, pre-processed data, and simulated data, quantifying statistical information and efficiency in scientific studies, particularly for genetic and environmental problems, statistical principles and foundational issues, such as multi-party inferences, the theory of ignorance, and the interplay between Bayesian and frequentist perspectives, effective deterministic and stochastic algorithms for Bayesian and likelihood computation and Bayesian inference, ranking and mapping.
Prof Meng received numerous awards including the 2001 COPSS (Committee of Presidents of Statistical Societies) Award, and also the 2003 Distinguished Achievement Award from International Chinese Statistics Association (ICSA), and was elected fellow by the Institute of Mathematical Statistics (IMS) in 1997 and by the American Statistical Association (ASA) in 2004. Most recently, he was elected the 2010 Medallion Lecturer of IMS, 2011 Mosteller Statistician of the Year by ASA Boston Chapter, the recipient of a 2011 Distinguished Alumni Award by Fudan University, and a recipient of the inaugural (2012) PL Hsu Award for distinguished achievements in research and education by a statistician under the age of fifty.
Duration: 87 min.
Appears in Series:8.15:3 - Audio-visual Materials
Videos for Public -- Distinguished Lectures