Statistics at UC Berkeley: We are a community engaged in research and education in probability and statistics. In addition to developing fundamental theory and methodology, we are actively involved in statistical problems that arise in such diverse fields as molecular biology, geophysics, astronomy, AIDS research, neurophysiology, sociology, political science, education, demography, and the U.S. Census. We have forged strong interdisciplinary links with other departments and areas of study, particularly biostatistics, mathematics, computer science, and biology, and actively seek to recruit graduate students and faculty who can help to build and maintain such links. We also offer a statistical consulting service each semester.
Statistics at UC Berkeley
Oct 5, 2016
Somayeh Sojoudi, Berkeley
This talk will explore the graphical LASSO and the associated SDP relaxation, deriving some simple conditions under which computational complexity can be managed. It will also introduce a circuit model that can be used as a platform for testing different statistical procedures.
Seminar 217, Risk Management: Machines Learning Justice: A New Approach to the Problems of Inconsistency and Bias in Adjudication
Ryan Copus and Hannah Laqueur, UC Berkeley (Speaker - Featured)
Abstract: We offer a two-step algorithmic approach to the problems of inconsistency and bias in legal decision making. First, we propose a new tool for reducing inconsistency: Judgmental Bootstrapping Models (“JBMs”) built with machine learning methods.
Yevgeniy Kovchegov, Oregon State University
The tree self-similarities based on Horton ordering and Tokunaga indexing were recently established for a variety of stochastic processes including the Kingman coalescent trees, level set trees for space-homogeneous stochastic processes, and iid time series. They were conjectured for the trees generated by the multiplicative and additive coalescent processes, and a number of other stochastic...
Chao Gao, Department of Statistics, University of Chicago
Abstract: I am going to give the motivation by discussing a simple location estimation problem under Huber’s contamination model. I will argue that Huber’s contamination model is a much better framework for robust statistics than Hampel’s breakdown point. Under Huber’s framework, Tukey’s location estimator is shown to be superior than the naive coordinate median. For the robust covariance matrix...
Gregory Schehr, Université de Paris-Sud
I will discuss extremal properties (the maximum and the time at which it occurs) of the top path of N non-intersecting Brownian bridges and excursions. In the case of excursions, I will show that, in the large N limit, the joint probability density function of the maximum and of its position can be described in terms of Painlevé transcendents. I will finally discuss the relevance of these results...