Abstract:The stochastic partial differential equation (SPDE) approach is widely used for modeling large spatial datasets. It is based on representing a Gaussian random field u on Rd as the solution of an elliptic SPDE Lβu=W where L is a second-order differential operator, 2β∈N is a positive parameter that controls the smoothness of u and W is Gaussian white noise. A few approaches have been suggested in the literature to extend the approach to allow for any smoothness parameter satisfying β>d/4 . Even though those approaches work well for simulating SPDEs with general smoothness, they are less suitable for Bayesian inference since they do not provide approximations which are Gaussian Markov random fields (GMRFs) as in the original SPDE approach. We address this issue by proposing a new method based on approximating the covariance operator L−2β of the Gaussian field u by a finite element method combined with a rational approximation of the fractional power. This results in a numerically stable GMRF approximation which can be combined with the integrated nested Laplace approximation (INLA) method for fast Bayesian inference. A rigorous convergence analysis of the method is performed and the accuracy of the method is investigated with simulated data. Finally, we illustrate the approach and corresponding implementation in the R package rSPDE via an application to precipitation data which is analyzed by combining the rSPDE package with the R-INLA software for full Bayesian inference. Supplementary materials for this article are available online.

Parameter estimation for second-order SPDEs in multiple space dimensions

Parameter estimation for SPDEs based on discrete observations in time and space

Parametric estimation for linear parabolic SPDEs in two space dimensions based on temporal and spatial increments

On Asymptotic Properties of the Parameter Estimator for a Type of SPDE

Optimal parameter estimation for linear SPDEs from multiple measurements

Non-parametric estimation of the reaction term in semi-linear SPDEs with spatial ergodicity

Parameter estimation in hyperbolic linear SPDEs from multiple measurements

Estimation for the damping factor of the driving process of an SPDE in two space dimensions

Small diffusivity asymptotics for a linear parabolic SPDE in two space dimensions

Small dispersion asymptotics for an SPDE in two space dimensions using triple increments

Parameter Estimation for SPDEs Driven by Cylindrical Stable Processes

Parameter estimations for SPDEs with multiplicative fractional noise

Nonparametric calibration for stochastic reaction-diffusion equations based on discrete observations

Parameter Estimation in Nonlinear Multivariate Stochastic Differential Equations Based on Splitting Schemes

Parameter Estimation in Stochastic Differential Equations

Covariance–Based Rational Approximations of Fractional SPDEs for Computationally Efficient Bayesian Inference

Parameter estimation for ergodic linear SDEs from partial and discrete observations

Parameter Estimation for Stochastic Partial Differential Equations of Second Order

Nonparametric velocity estimation in stochastic convection-diffusion equations from multiple local measurements

Bayesian parameter inference for partially observed stochastic volterra equations

Parameter estimation of ODE's via nonparametric estimators