Introduction

MCMCLib is a lightweight C++ library of Markov Chain Monte Carlo (MCMC) methods.

Features:

• A C++11/14/17 library of well-known MCMC algorithms.

• Parallelized samplers designed for multi-modal distributions, including:

  • Differential Evolution (DE); and

  • Adaptive Equi-Energy Sampler (AEES).

• For fast and efficient matrix-based computation, MCMCib supports the following templated linear algebra libraries:

  • Armadillo

  • Eigen (version >= 3.4.0)

• Automatic differentiation functionality is available through use of the Autodiff library

• OpenMP-accelerated algorithms for parallel computation.

• Straightforward linking with parallelized BLAS libraries, such as OpenBLAS.

• Available as a header-only library, or as a compiled shared library.

• Released under a permissive, non-GPL license.

Author: Keith O’Hara

License: Apache Version 2.0

---

Installation

The library can be installed on Unix-alike systems via the standard ./configure && make method.

See the installation page for detailed instructions.

Algorithms

A list of currently available algorithms includes:

• Adaptive Equi-Energy Sampler (AEES)

• Differential Evolution (DE-MCMC)

• Hamiltonian Monte Carlo (HMC)

• Metropolis-adjusted Langevin algorithm (MALA)

• No-U-Turn Sampler (NUTS)

• Random Walk Metropolis-Hastings (RWMH)

• Riemannian Manifold Hamiltonian Monte Carlo (RM-HMC)

---

Contents

Guide

• Installation
  • Requirements
  • Installation Method 1: Shared Library
  • Installation Method 2: Header-only Library
• Examples and Tests
  • API
  • Example
  • Test suite
• MCMC Settings
  • Main
  • By Algorithm
• Automatic Differentiation
  • Example

Algorithms

• Adaptive Equi-Energy Sampler
• Differential Evolution
• Hamiltonian Monte Carlo
• Metropolis-adjusted Langevin Algorithm
• No-U-Turn Sampler
• Random Walk Metropolis-Hastings
• Riemannian Manifold HMC

Appendix

• Box Constraints