Accurate Monte Carlo estimation of very small \(p\)-values in Markov chains (Q1297844)

The author proposed an algorithm that enable one to calculate very small \(p\)-values of, for instance, tail probabilities connected with probability density \(p(x)\) for some complicated random models that can be realized as a Markov process. The main idea is to represent a probability density \(p(x)\) as the equilibrium state for a finite time homogeneous discrete Markov chain. Such an approach is illustrated by three examples, particularly, including the protein threading model.