SPDE scaling limits of an Markov chain Montecarlo algorithm
Seminar Room 1, Newton Institute
AbstractI will discuss how a simple random walk metropolis algorithm converges to an SPDE as the dimension of the sample space goes to infinity. I will discuss how this the limiting SPDE gives insight into how one should tune the algorithm to obtain an asymptotically optimal mixing rate. This is joint work with Andrew Stuart and Natesh Pialli.
If it doesn't, something may have gone wrong with our embedded player.
We'll get it fixed as soon as possible.