Department of Statistics
Persi Diaconis is the Mary V. Sunseri Professor of Statistics and Mathematics at Stanford University. Before joining the faculty at Stanford University, he was a professor of mathematics at both Harvard University and Cornell University. He is a MacArthur Fellow and a member of the National Academy of Sciences. He has received honorary degrees from the University of Chicago, Universite Paul Sabatier (Toulouse), Queen Mary University of London, and Athens (Greece) University of Economics and Business. He was the IMS Wald Lecturer, the SIAM Von Neumann Lecturer, and the President of IMS. He has supervised more than 40 PhD students. Dr. Diaconis is particularly known for tackling mathematical problems involving randomness and randomization such as coin flipping and shuffling playing cards. He is a former professional magician.
(September 13, 2010)
The Search for Randomness
I will take a close look at some of our most primitive images of randomness: flipping a coin, shuffling cards, and rolling a roulette ball. In each case, while things can be made random, usually we are lazy. A bit of math and experimentation shows real-world instances can be quite non-random. Philosophical consequences and applications to the use (and misuse) of statistical models are given.
(September 14, 2010)
Statistical Analysis of Markov Chains
In protein folding, chemistry, and social science, time series are modeled as Markov chains. A Bayesian approach to testing and estimation calls for tractable priors. Developing these for, e.g., reversible chains or continuous stat spaces, needs new mathematics. The tools of reinforced random walk in many variations are helpful. This is joint work with Silke Rolles and Wai-Wai Liu.