Dynamic importance sampling for uniformly recurrent Markov chains (Q1774208)
From MaRDI portal
scientific article
| Language | Label | Description | Also known as |
|---|---|---|---|
| English | Dynamic importance sampling for uniformly recurrent Markov chains |
scientific article |
Statements
Dynamic importance sampling for uniformly recurrent Markov chains (English)
0 references
29 April 2005
0 references
A dynamic important sampling in the setting of uniformly recurrent Markov chain is suggested for reducing the variance of estimating the rare probability by Monte Carlo, where the reference measure used for rejection principle is time varying. The existence of asymptotically optimal scheme is shown based on a control theory approach. Numerical examples are illustrated in contrast to the standard scheme.
0 references
asymptotic optimality
0 references
important sampling
0 references
Monte Carlo simulation
0 references
rare event
0 references
weak convergence
0 references
0 references
0 references
0 references
0 references
0 references
