Comparison of hit-and-run, slice sampling and random walk Metropolis
arXiv:1505.00579 · doi:10.1017/jpr.2018.78
Abstract
Different Markov chains can be used for approximate sampling of a distribution given by an unnormalized density function with respect to the Lebesgue measure. The hit-and-run, (hybrid) slice sampler and random walk Metropolis algorithm are popular tools to simulate such Markov chains. We develop a general approach to compare the efficiency of these sampling procedures by the use of a partial ordering of their Markov operators, the covariance ordering. In particular, we show that the hit-and-run and the simple slice sampler are more efficient than a hybrid slice sampler based on hit-and-run which, itself, is more efficient than a (lazy) random walk Metropolis algorithm.
18 pages, Accepted for publication by the Applied Probability Trust (http://www.appliedprobability.org) in J. Appl. Prob
References in corpus (7)
- Renewal theory and computable convergence rates for geometrically ergodic Markov chains
- Positivity of hit-and-run and related algorithms
- Optimal scaling of random walk Metropolis algorithms with discontinuous target densities
- Rapid mixing of Swendsen-Wang dynamics in two dimensions
- Hit-and-run for numerical integration
- Rapid mixing of Swendsen-Wang and single-bond dynamics in two dimensions
- Convergence of hybrid slice sampling via spectral gap