The STOCHASTICA SNIP seminar will be held online every two weeks starting on January 14. Talks of 30-45 minutes will be followed by 15-30 minutes of open discussion. Initially, seminars will be focused on introducing new problems and applications, to stimulate further discussions and ideas within individual working groups.  

The sixth seminar was given by Prof Michael V. Tretyakov (University of Nottingham, UK), and chaired by Prof. Mireille Bossy (Leader, Working Group 3).

Switching diffusions for sampling from mixture distributions

We propose to use stochastic differential equations with state-dependent switching rates (SDEwS) for sampling from finite mixture distributions.

We consider an Euler scheme with constant time step for SDEwS. We prove its  weak first-order convergence and also in the ergodic limit. We present numerical experiments that illustrate the use of SDEwS for sampling from mixture distributions and confirm the theoretical results.