Variational State and Parameter Estimation

This paper considers the problem of computing Bayesian estimates both states and model parameters for nonlinear state-space models. Generally, this does not have a tractable solution approximations must be utilised. In work, variational approach is used to provide an assumed density which approximates desired, intractable, distribution. The deterministic results in optimisation standard form. Due parametrisation selected first second order derivatives are readily available allows efficient solutions. proposed method compared against state-of-the-art Hamiltonian Monte Carlo two numerical examples.