Statistics Seminar
Date/Time: | Monday, 28 Mar 2016 from 4:10 pm to 5:00 pm |
---|---|
Location: | Snedecor 3105 |
Cost: | Free |
URL: | stat.iastate.edu |
Contact: | Denise Riker |
Phone: | 515-294-1076 |
Channel: | College of Liberal Arts and Sciences |
Categories: | Lectures |
Actions: | Download iCal/vCal | Email Reminder |
The classical Glivenko-Cantelli theorem says that for iid rv the empirical cdf converges to the true cdf uniformly on R w.p.1. In this talk we extend this to delayed regenerative sequences, delayed stationary sequences and delayed exchangeable sequences. Then we apply this to irreducible Markov chains admitting a stationary probability distribution with either a countable state space or Harris recurrent Markov chains.
This is of relevance to MCMC methodology. A key tool used is a generalisation of Polya's theorem on convergence of cdfs.
(joint work with Vivek Roy of ISU Statistics)