DMSS2006: The International Workshop on Data-Mining and Statistical
September 25-26, 2006, Century Royal Hotel, Sapporo, Japan
Random Sampling via Markov Chain
Shuji Kijima, Tomomi Matsui
The Markov chain Monte Carlo method is based on a simple idea, and works powerfully for samplinghard objects. Recently, it appears in many areas such as statistical physics, economics, statistics, and so on. There are some technical points when to use it practically, and in this talk, we focus on a topic of convergence rate of Markov chains, with some examples such as contingency tables, Ising model, and permanent. We also talk about perfect sampling.
Shuji Kijima's HP