Yale University
Department of Statistics
Seminar

Monday, February 26, 2001


Oliver Johnson
Christ's College and Statistical Laboratory
University of Cambridge


Information Theoretic Central Limit Theorems

We present new results concerning information-theoretic convergence in the Central Limit Theorem. In joint work with Barron, we present a simplified proof, which also provides an explicit rate of convergence, using so-called Poincare inequalities. We also present techniques which can provide a proof in certain dependent cases, under either a Rosenblatt-style mixing condition, or for FKG systems.





Seminar to be held in Room 107, 24 Hillhouse Avenue at 4:15 pm