JuliaBUGS.jl

JuliaBUGS.jl is a modern, high-performance implementation of the BUGS probabilistic programming language in Julia. It brings the familiar BUGS modelling syntax into the Julia ecosystem, enabling Bayesian inference with speed, flexibility, and seamless integration with the scientific computing ecosystem in Julia.

---

Why JuliaBUGS?

• BUGS syntax — write models using standard BUGS notation.

• High performance — leverage Julia’s speed and just-in-time (JIT) compilation.

• Interoperability with Julia ecosystem — works smoothly with MCMC algorithms from Turing.jl and the broader Julia PPL ecosystem.

---

Quick Start

A simple example model:

using JuliaBUGS, Random, AbstractMCMC

normal_model = JuliaBUGS.@bugs"""
model {
  for (i in 1:N) {
    y[i] ~ dnorm(mu, tau)
  }
  mu ~ dnorm(0, 0.001)
  tau ~ dgamma(0.1, 0.1)
}
"""

# The model definition is callable: pass data to construct a model.
posterior = normal_model((; N = 10, y = randn(10)))
rng, sampler = Random.MersenneTwister(123), JuliaBUGS.IndependentMH()

chain = AbstractMCMC.sample(rng, posterior, sampler, 1000)

For a complete walkthrough, see the example.

---

Related Tools

• DoodleBUGS — a browser-based interface for drawing BUGS models.
• RJuliaBUGS — an R interface to JuliaBUGS.

For alternative BUGS-family tools, see JAGS and Nimble.