专业描述

The Department of Statistics of the University was established in 1949 to conduct research into advanced statistics and probability, to work with others in the application of statistics to investigations in the natural and social sciences, and to teach probability and statistical theory and practice on the undergraduate and graduate levels. From its beginning, the Department has been recognized for the high quality of its faculty and the diversity of its interests. Some of the most important and influential texts and monographs in statistics and probability of the past forty years have been authored by former faculty members of our Department. These include Ergodic Theory and Information, Convergence of Probability Measures, and Probability and Measure by Patrick Billingsley; Inference and Disputed Authorship: The Federalist, an application of Bayesian methods to fix the authorship of the Federalist Papers, by David L. Wallace and Frederick Mosteller; and The Foundations of Statistics, a famous analysis of fundamental problems by Leonard J. Savage. Current members of our faculty have written definitive works in a variety of areas of current research interest. These include Generalized Linear Models, an influential monograph that extends the scope of linear models greatly, including to models for discrete data, by Peter McCullagh and John Nelder; Tensor Methods in Statistics, a monograph on methods for making complex multivariate calculations, by Peter McCullagh; Elements of Statistical Computing: Numerical Computation, a far-ranging text on numerical methods for statistics by Ronald A. Thisted; The History of Statistics: The Measurement of Uncertainty Before 1900, and Statistics on the Table, accounts by Stephen M. Stigler of the historical development of the field of mathematical statistics; Interpolation of Spatial Data: Some Theory for Kriging, a monograph providing a sound mathematical basis for understanding the behavior of a popular methodology for prediction of spatial processes by Michael L. Stein; Michael J. Wichura recently published a fundamental graduate text, The Coordinate Free Approach to Linear Models; Lars Peter Hansen (with Thomas Sargent) recently published Robustness, an adaptation of robust control techniques to mis-specification problems in economics; Kirk Wolter's 2nd edition of his classic Introduction to Variance Estimation was recently issued; 2D Object Detection and Recognition, Models, Algorithms, and Networks provides a state-of-the-art account of statistical methods in computer vision by Yali Amit; and Greg Lawler's Random Walk: A Modern Introduction has just been published. Faculty members have contributed many articles to books and journals in theoretical and applied statistics, biophysics, chemistry, mathematics, geophysics, astronomy, genetics, neuroscience, bacteriology, biometry, public health, machine learning, artificial intelligence, imaging, psychology, sociology, medicine, law, history of science, education, and business. Members of the department have at various times edited the four leading American or international journals of probability and statistics. Steve Lalley and Greg Lawler were previous editors of the Annals of Probability, the foremost research journal in the theory of probability. Several faculty members have been president of one or both of the two leading societies. Peter McCullagh, a leader in the development of generalized linear models, is a Fellow of the Royal Society. Stephen Stigler served recently as the President of the International Statistical Institute and was recently appointed to the Royal Academy of Belgium. Michael Stein is a leading expert in spatial and environmental statistics. Per Mykland uses his expertise in martingale theory and stochastic calculus to better understand financial markets. Yali Amit is developing fundamentally new approaches to computer vision and works on neural models for memory and recognition. Mary Sara McPeek studies genetic association and is a leader in understanding the statistical impact of family relations between sampled individuals. Matthew Stephens studies genetic association, sequencing and population genetics and is a leader in the application of modern Bayesian methods in genetics. Dan Nicolae is co-chair of a national asthma genetics consortium and studies genetic and environmental factors affecting human disease. Wei-Biao Wu is developing novel mathematical approaches to the analysis of time series. The department is expanding its horizons into various fields of computational and applied mathematics: John Reinitz is a mathematical biologist who studies mathematical models for the development of genetic expression patterns. John Lafferty, who introduced conditional random fields in machine learning, currently studies the statistical properties of high dimensional sparse models. Nicolas Brunel works in computational neuroscience studying the dynamic properties of models for individual neurons and their ensembles, and Mihai Anitescu, who is joint with Argonne National Lab, studies nonconvex constrained optimization with application to large-scale problems in the energy domain.