Abstract

In this chapter we consider how approximate Bayesian computation (ABC) methods can be used in estimating the (unknown) parameters θ of a parametric model (stochastic process) ℳ , where in our case the parametric model ℳ will be a stochastic epidemic process. We assume that ℳ gives rise to realizations X with the observed data x ∗ being one such realization. The foundations for ABC, along with all Bayesian statistics, is Bayes’ Theorem which states that 10.1 π (   θ   | x ∗ )   =   π ( x ∗ |   θ   ) π (   θ   ) π ( x ∗ ) ,