Julia
-
CBPFCouplingPaperCodes: Implementation of coupled conditional backward sampling particle filter algorithms and experiments in arXiv:2312.17572
-
GenericSSMs.jl: Package (by Santeri Karppinen) for generic particle filtering and particle Markov chain Monte Carlo
-
cpf-bbs: Codes (with Santeri Karppinen) for experiments in the paper doi:10.1080/10618600.2023.2231514/arXiv:2205.13898, about 'bridge backward sampling' generalisation of backward/ancestor sampling, which avoids degeneracy with models having weakly informative observations and stiff dynamics.
-
weakly-informative-resampling-codes: Codes for experiments in the paper doi:10.1214/22-AOS2222/arXiv:2203.10037, about performance of various resampling methods in the weakly informative regime.
- Resamplings.jl: Package (with Santeri Karppinen) which implements a number of unbiased resampling algorithms,which can be used easily within a particle filter (sequential Monte Carlo) algorithm.
-
tassu-filtering: Codes (by Santeri Karppinen) which implement the method for estimating animal territories and which was used in the experiments of the related paper doi:10.1016/j.ecolmodel.2022.110101.
- cpf-diff-init: Codes (by Santeri Karppinen) which implement the conditional
particle filter with diffuse initialisation, and the experiments of
the related paper doi:10.1007/s11222-020-09975-1/arXiv:2006.14877.
- AdaptiveMCMC.jl: Package implementing random-walk based adaptive MCMC as discussed in the related paper doi:10.1002/9781118445112.stat08286/Author version
algorithms
- AdaptiveParticleMCMC.jl: Package implementing particle MCMC algorithms with adaptive
proposals
- CoupledConditionalSMC.jl: Package (by Anthony Lee)
implementing coupled conditional particle filters (for unbiased smoothing)
suggested in doi:10.1214/19-AOS1922/arXiv:1806.05852.
- AdaptiveToleranceABC_MCMC.jl: Package which implements an adaptive tolerance ABC-MCMC with
post-correction, as described in the related paper
doi:10.1093/biomet/asz078/arXiv:1902.00412. See also
abc-mcmc
for the codes of the experiments in the paper.
R
Python
- ram.py: Simple pure Python (& NumPy) implementation of the
RAM algorithm
Matlab
- apt-codes: Matlab codes implementing the adaptive parallel tempering algorithm(s) on the experiments described in the related paper
- ram_demo.m: Simple Matlab implementation of the RAM
C/C++
- unbiased-mlmc: C++ implementation of unbiased multilevel Monte Carlo estimators for diffusion expectations, as described in the paper
- grapham: C
implementation of several adaptive random-walk Metropolis
algorithms on graphical models (or Bayesian networks), with models
defined in Lua
(and optionally partly in C)