# Quasi Markov Chain Monte Carlo Methods

@article{Schwedes2018QuasiMC, title={Quasi Markov Chain Monte Carlo Methods}, author={Tobias Schwedes and Ben Calderhead}, journal={arXiv: Statistics Theory}, year={2018} }

Quasi-Monte Carlo (QMC) methods for estimating integrals are attractive since the resulting estimators typically converge at a faster rate than pseudo-random Monte Carlo. However, they can be difficult to set up on arbitrary posterior densities within the Bayesian framework, in particular for inverse problems. We introduce a general parallel Markov chain Monte Carlo (MCMC) framework, for which we prove a law of large numbers and a central limit theorem. In that context, non-reversible… Expand

#### Figures and Tables from this paper

#### 3 Citations

Markov chain Monte Carlo importance samplers for Bayesian models with intractable likelihoods

- Mathematics, Computer Science
- 2019

Convergence of the estimator is verified to hold under regularity assumptions which do not require that the diffusion can be simulated exactly, and an adaptive MCMC approach to deal with the selection of a suitably large tolerance is suggested. Expand

Conditional sequential Monte Carlo in high dimensions

- Mathematics
- 2021

The iterated conditional sequential Monte Carlo (i-CSMC) algorithm from Andrieu, Doucet and Holenstein (2010) is an MCMC approach for efficiently sampling from the joint posterior distribution of the… Expand

MetFlow: A New Efficient Method for Bridging the Gap between Markov Chain Monte Carlo and Variational Inference

- Computer Science, Mathematics
- ArXiv
- 2020

A new computationally efficient method to combine Variational Inference (VI) with Markov Chain Monte Carlo (MCMC) is proposed, which is amenable to the reparametrization trick and does not rely on computationally expensive reverse kernels. Expand

#### References

SHOWING 1-10 OF 101 REFERENCES

A Randomized Quasi-Monte Carlo Simulation Method for Markov Chains

- Computer Science, Mathematics
- Oper. Res.
- 2008

A randomized quasi-Monte Carlo method for the simulation of Markov chains up to a random (and possibly unbounded) stopping time that proves bounds on the convergence rate of the worst-case error and of the variance for special situations where the state space of the chain is a subset of the real numbers. Expand

A general construction for parallelizing Metropolis−Hastings algorithms

- Computer Science, Medicine
- Proceedings of the National Academy of Sciences
- 2014

This paper proposes a natural generalization of the Metropolis−Hastings algorithm that allows for parallelizing a single chain using existing MCMC methods, and shows how it allows for a principled way of using every integration step within Hamiltonian Monte Carlo methods. Expand

Consistency of Markov chain quasi-Monte Carlo on continuous state spaces

- Mathematics
- 2011

The random numbers driving Markov chain Monte Carlo (MCMC) simulation are usually modeled as independent U(0, 1) random variables. Tribble [Markov chain Monte Carlo algorithms using completely… Expand

Monte Carlo Sampling Methods Using Markov Chains and Their Applications

- Mathematics
- 1970

SUMMARY A generalization of the sampling method introduced by Metropolis et al. (1953) is presented along with an exposition of the relevant theory, techniques of application and methods and… Expand

Riemann manifold Langevin and Hamiltonian Monte Carlo methods

- Mathematics
- 2011

The paper proposes Metropolis adjusted Langevin and Hamiltonian Monte Carlo sampling methods defined on the Riemann manifold to resolve the shortcomings of existing Monte Carlo algorithms when… Expand

Multidimensional variation for quasi-Monte Carlo

- Mathematics
- 2004

This paper collects together some properties of multidimensional definitions of the total variation of a real valued function. The subject has been studied for a long time. Many of the results… Expand

Randomized Quasi-Monte Carlo Simulation of Markov Chains with an Ordered State Space

- Mathematics
- 2006

We study a randomized quasi-Monte Carlo method for estimating the state distribution at each step of a Markov chain with totally ordered (discrete or continuous) state space. The number of steps in… Expand

New Inputs and Methods for Markov Chain Quasi-Monte Carlo

- Computer Science
- 2012

Some new constructions of points, fully equidistributed LFSRs, are presented, which are small enough that the entire point set can be used in a Monte Carlo calculation. Expand

When Are Quasi-Monte Carlo Algorithms Efficient for High Dimensional Integrals?

- Computer Science, Mathematics
- J. Complex.
- 1998

It is proved that the minimalworst case error of quasi-Monte Carlo algorithms does not depend on the dimensiondiff the sum of the weights is finite, and the minimal number of function values in the worst case setting needed to reduce the initial error by ? is bounded byC??p, where the exponentp? 1, 2], andCdepends exponentially on thesum of weights. Expand

Pseudorandom numbers for modelling Markov chains

- Mathematics
- 1967

Abstract THE calculation of space averages in problems of statistical physics can be reduced to the Monte Carlo simulation of the mathematical expectations of the corresponding quantities based on… Expand