This repository proposes several fonctions used in [1]. In this paper, we propose a new approximate Bayesian computation (ABC) algorithm that we compared with three other algorithms: the Population Monte Carlo algorithm [2], the Sequential Monte Carlo algorithm [3] and the Replenishement Sequential Monte Carlo algorithm [4]. We compare them using the toy example studied in [5]
This repository contains an implementation of the four ABC algorithms described in the paper,
- APMC: Adaptive population Monte Carlo [1]
- PMC: Population Monte Carlo [2]
- SMC: Sequential Monte Carlo [3]
- RSMC: Replenishement Sequential Monte Carlo [4]
and an implementation of the toy model,
- ToyEx: Toy example proposed by [5]
- L2: Compute the L2 distance between a simulated (weighted particles) and the true posterior density for the toy example proposed by [5]
- sample: Sample x particles from a weighted distribution
[1] Lenormand et al. (2013) Adaptive approximate Bayesian computation for complex models. Computational Statistics 28, 2777-2796.
[2] Beaumont et al. (2009) Adaptive approximate Bayesian computation. Biometrika 96, 983–990.
[3] Del Moral et al. (2012) An adaptive sequential Monte Carlo method for approximateBayesian computation. Statistics and Compututing 22, 1009–1020.
[4] Drovandi & Pettitt (2011) Estimation of parameters for macroparasite population evolution usingapproximate Bayesian computation. Biometrics 67, 225–233.
[5] Sisson et al. (2007) Sequential Monte Carlo without likelihoods. PNAS 104, 1760-1765.
If you use this code, please cite:
Lenormand M, Jabot F & Deffuant G (2013) Adaptive approximate Bayesian computation for complex models. Computational Statistics 28, 2777-2796.
If you need help, find a bug, want to give me advice or feedback, please contact me! You can reach me at maxime.lenormand[at]inrae.fr