In this paper we define and study new directional Metropolis–Hastings algorithms that propose states in hyperplanes. Each iteration in directional Metropolis–Hastings algorithms consist of three steps. First a direction is sampled by an auxiliary variable. Then a potential new state is proposed in the subspace defined by this direction and the current state. Lastly, the potential new state is a...

The word ‘Metropolis’ is a shorter version of the word ‘meterpolis’, itself derived from the Greek word ‘meter’ (mother) and polis (city, town). It is identified as the ‘center city’ which is more developed than other cities in terms of culture and economy (1). Hence, the metropolis is a depiction of hybridity and multi-layering; and while it is related to the city, it is not solely derived fro...

Your use of the JSTOR archive indicates your acceptance of JSTOR's Terms and Conditions of Use, available at http://www.jstor.org/about/terms.html. JSTOR's Terms and Conditions of Use provides, in part, that unless you have obtained prior permission, you may not download an entire issue of a journal or multiple copies of articles, and you may use content in the JSTOR archive only for your perso...

For SU(2) lattice gauge theory with the fundamental-adjoint action an efficient heat-bath algorithm is not known so that one had to rely on Metropolis simulations supplemented by overrelaxation. Implementing a novel biased Metropolis-heat-bath algorithm for this model, we find improvement factors in the range 1.45 to 2.06 over conventionally optimized Metropolis simulations. If one optimizes fu...

The Metropolis process is an extremely general recipe for constructing a Markov chain which has any desired stationary distribution π on a finite set Ω. Moreover, this distribution can be specified just by a weight function w : Ω → R, so that π(x) = w(x) Z where Z is an unknown normalizing factor. The Metropolis process is named after one of its inventors [MR+53]. To specify the Metropolis proc...

The random walk Metropolis algorithm is a simple Markov chain Monte Carlo scheme which is frequently used in Bayesian statistical problems. We propose a guided walk Metropolis algorithm which suppresses some of the random walk behavior in the Markov chain. This alternative algorithm is no harder to implement than the random walk Metropolis algorithm, but empirical studies show that it performs ...

A hybrid algorithm is proposed for pure SU(N) lattice gauge theory based on Genetic Algorithms (GA)s and the Metropolis method. We apply the hybrid GA to pure SU(2) gauge theory on a 2-dimensional lattice and find the action per plaquette and Wilson loops being consistent with those given by the Metropolis and Heatbath methods. The thermalization of this newly proposed Hybrid GA is quite faster...

The waste-recycling Monte Carlo (WRMC) algorithm introduced by physicists is a modification of the (multi-proposal) Metropolis–Hastings algorithm, which makes use of all the proposals in the empirical mean, whereas the standard (multi-proposal) Metropolis–Hastings algorithm uses only the accepted proposals. In this paper we extend the WRMC algorithm to a general control variate technique and ex...