Abstract:In this article we consider static Bayesian parameter estimation for partially observed diffusions that are discretely observed. We work under the assumption that one must resort to discretizing the underlying diffusion process, for instance, using the Euler-Maruyama method. Given this assumption, we show how one can use Markov chain Monte Carlo (MCMC) and particularly particle MCMC [C. Andrieu, A. Doucet, and R. Holenstein, T. R. Stat. Soc. Ser. B Stat. Methodol., 72 (2010), 269-342] to implement a new approximation of the multilevel (ML) Monte Carlo (MC) collapsing sum identity. Our approach comprises constructing an approximate coupling of the posterior density of the joint distribution over parameter and hidden variables at two different discretization levels and then correcting by an importance sampling method. The variance of the weights are independent of the length of the observed data set. The utility of such a method is that, for a prescribed level of mean square error, the cost of this MLMC method is provably less than i.i.d. sampling from the posterior associated to the most precise discretization. However the method here comprises using only known and efficient simulation methodologies. The theoretical results are illustrated by inference of the parameters of two prototypical processes given noisy partial observations of the process: the first is an Ornstein Uhlenbeck process and the second is a more general Langevin equation.

On nonparametric estimation of the interaction function in particle system models

Kinetic interacting particle system: parameter estimation from complete and partial discrete observations

Optimal minimax rate of learning interaction kernels

A Method of Moments Estimator for Interacting Particle Systems and their Mean Field Limit

Polynomial rates via deconvolution for nonparametric estimation in McKean-Vlasov SDEs

Dose-Response Curve Estimation: A Semiparametric Mixture Approach

Non-parametric estimates for graphon mean-field particle systems

A Few Interactions Improve Distributed Nonparametric Estimation, Optimally

Eigenfunction martingale estimators for interacting particle systems and their mean field limit

Nonparametric estimation via partial derivatives

Bayesian Static Parameter Estimation for Partially Observed Diffusions Via Multilevel Monte Carlo.

Convergence Rate for Moderate Interaction particles and Application to Mean Field Games

Nonparametric Estimation and Testing of Interaction in Additive Models

Nonparametric Bayesian inference of discretely observed diffusions

Nonparametric estimation for a class of Lévy processes

Inferring interaction potentials from stochastic particle trajectories

On the Identifiablility of Nonlocal Interaction Kernels in First-Order Systems of Interacting Particles on Riemannian Manifolds

Inference of interaction kernels in mean-field models of opinion dynamics

Interacting Particle Systems on Networks: joint inference of the network and the interaction kernel