A synthetic likelihood-based Laplace approximation for efficient design of biological processes

Abstract

Complex models used to describe biological processes in epidemiology and ecology often have computationally intractable or expensive likelihoods. This poses significant challenges in terms of Bayesian inference but more significantly in the design of experiments. Bayesian designs are found by maximising the expectation of a utility function over a design space, and typically this requires sampling from or approximating a large number of posterior distributions. This renders approaches adopted in inference computationally infeasible to implement in design. Consequently, optimal design in such fields has been limited to a small number of dimensions or a restricted range of utility functions. To overcome such limitations, we propose a synthetic likelihood-based Laplace approximation for approximating utility functions for models with intractable likelihoods. As will be seen, the proposed approximation is flexible in that a wide range of utility functions can be considered, and remains computationally efficient in high dimensions. To explore the validity of this approximation, an illustrative example from epidemiology is considered. Then, our approach is used to design experiments with a relatively large number of observations in two motivating applications from epidemiology and ecology.