Approximate Bayesian computation: a simulation based approach to inference

There is a large class of stochastic models for which we can simulate observations from the model, but for which the likelihood function is unknown. Without knowledge of the likelihood function standa