We gratefully acknowledge support from
the Simons Foundation and member institutions.
Full-text links:


Current browse context:


Change to browse by:

References & Citations


(what is this?)
CiteULike logo BibSonomy logo Mendeley logo del.icio.us logo Digg logo Reddit logo ScienceWISE logo

Statistics > Computation

Title: Metropolis-Hastings algorithms with autoregressive proposals, and a few examples

Abstract: We analyse computational efficiency of Metropolis-Hastings algorithms with stochastic AR(1) process proposals. These proposals include, as a subclass, discretized Langevin diffusion (e.g. MALA) and discretized Hamiltonian dynamics (e.g. HMC).
We derive expressions for the expected acceptance rate and expected jump size for MCMC methods with general stochastic AR(1) process proposals for the case where the target distribution is absolutely continuous with respect to a Gaussian and the covariance of the Gaussian is allowed to have off-diagonal terms. This allows us to extend what is known about several MCMC methods as well as determining the efficiency of new MCMC methods of this type. In the special case of Hybrid Monte Carlo, we can determine the optimal integration time and the effect of the choice of mass matrix.
By including the effect of Metropolis-Hastings we also extend results by Fox and Parker, who used matrix splitting techniques to analyse the performance and improve efficiency of stochastic AR(1) processes for sampling from Gaussian distributions.
Comments: 58 pages. Electronics Technical Reports No. 2016-1, ISSN 1172-496X, May 2016. arXiv admin note: substantial text overlap with arXiv:1501.03150
Subjects: Computation (stat.CO)
MSC classes: 60J22, 65Y20, 60J05, 62M05, 65C40, 68Q87, 68W20
Cite as: arXiv:1605.05441 [stat.CO]
  (or arXiv:1605.05441v2 [stat.CO] for this version)

Submission history

From: Richard Norton [view email]
[v1] Wed, 18 May 2016 05:08:48 GMT (144kb)
[v2] Thu, 19 May 2016 23:08:45 GMT (144kb)

Link back to: arXiv, form interface, contact.