Particle Methods for Stochastic Differential Equation Mixed Effects Models

Parameter inference for stochastic differential equation mixed effects models (SDEMEMs) is challenging. Analytical solutions these are rarely available, which means that the likelihood also intractable. In this case, exact (up to discretisation of equation) possible using particle MCMC methods. Although posterior targeted by methods, a naive implementation SDEMEMs can be highly inefficient. Our article develops three extensions approach exploit specific aspects and other advances such as correlated pseudo-marginal We compare methods on simulated data from tumour xenography study mice.