Adaptive approximate Bayesian computation

About

Sequential techniques can enhance the efficiency of the approximate Bayesian computation algorithm, as in Sisson et al.'s (2007) partial rejection control version. While this method is based upon the theoretical works of Del Moral et al. (2006), the application to approximate Bayesian computation results in a bias in the approximation to the posterior. An alternative version based on genuine importance sampling arguments bypasses this difficulty, in connection with the population Monte Carlo method of Cappe et al. (2004), and it includes an automatic scaling of the forward kernel. When applied to a population genetics example, it compares favourably with two other versions of the approximate algorithm.

Mark A. Beaumont, Jean-Marie Cornuet, Jean-Michel Marin, Christian P. Robert · 2008

Related benchmarks

Task	Dataset	Result
Simulation-Based Inference	SBIBM Gaussian Linear	C2ST0.8	19
Bayesian Inference	Bayesian Inverse Problems Task 1 1.0	C2ST72.6	17
Bayesian Inference	Bayesian Inverse Problems Task 2 1.0	C2ST79.4	17
Bayesian Inference	Bayesian Inverse Problems Task 3 1.0	C2ST96.3	17
Bayesian Inference	Bayesian Inverse Problems Task 5 1.0	C2ST66.3	17
Bayesian Inference	Bayesian Inverse Problems Task 4 1.0	C2ST66.4	17
Simulation-Based Inference	Gaussian Linear	Computation Time (s)0.25	8
Simulation-Based Inference	Gaussian Mixture	Computation Time (s)0.37	8
Simulation-Based Inference	Bernoulli GLM	Computation Time (s)4.88	8
Simulation-Based Inference	Two Moons	Computation Time (s)0.4	8