arXiv subjects

Daniel Simpson

Publications and source records attributed to Daniel Simpson.

45 records · Page 3Linked to original sources

arXiv2013

Specifying Gaussian Markov Random Fields with Incomplete Orthogonal Factorization using Givens Rotations

In this paper an approach for finding a sparse incomplete Cholesky factor through an incomplete orthogonal factorization with Givens rotations is discussed and applied to Gaussian Markov random fields (GMRFs). The incomplete Cholesky factor obtained from the incomplete orthogonal factorization is usually sparser than the commonly used Cholesky factor obtained through the standard Cholesky factorization. On the computational side, this approach can provide a sparser Cholesky factor, which gives a computationally more efficient representation of GMRFs. On the theoretical side, this approach is stable and robust and always returns a sparse Cholesky factor. Since this approach applies both to square matrices and to rectangle matrices, it works well not only on precision matrices for GMRFs but also when the GMRFs are conditioned on a subset of the variables or on observed data. Some common structures for precision matrices are tested in order to illustrate the usefulness of the approach. One drawback to this approach is that the incomplete orthogonal factorization is usually slower than the standard Cholesky factorization implemented in standard libraries and currently it can be slower to build the sparse Cholesky factor.

stat.CO↗

arXiv2013

Multivariate Gaussian Random Fields with Oscillating Covariance Functions using Systems of Stochastic Partial Differential Equations

In this paper we propose a new approach for constructing \emph{multivariate} Gaussian random fields (GRFs) with oscillating covariance functions through systems of stochastic partial differential equations (SPDEs). We discuss how to build systems of SPDEs that introduces oscillation characteristics in the covariance functions of the multivariate GRFs. By choosing different parametrization of the equations, some GRFs can be made with oscillating covariance functions but other fields can have Matérn covariance functions or close to Matérn covariance functions. The multivariate GRFs constructed by solving the systems of SPDEs automatically fulfill the hard requirement of nonnegative definiteness for the covariance functions. The approximate weak solutions to the systems of SPDEs are used to represent the multivariate GRFs by multivariate Gaussian \emph{Markov} random fields (GMRFs). Since the multivariate GMRFs have sparse precision matrices (inverse of the covariance matrices), numerical algorithms for sparse matrices can be applied to the precision matrices for sampling and inference. Thus from a computational point of view, the \emph{big-n} problem can be partially solved with these types of models. Another advantage of the method is that the oscillation in the covariance function can be controlled directly by the parameters in the system of SPDEs. We show how to use this proposed approach with simulated data and real data examples.

stat.ME↗

arXiv2013

Non-stationary Spatial Modelling with Applications to Spatial Prediction of Precipitation

A non-stationary spatial Gaussian random field (GRF) is described as the solution of an inhomogeneous stochastic partial differential equation (SPDE), where the covariance structure of the GRF is controlled by the coefficients in the SPDE. This allows for a flexible way to vary the covariance structure, where intuition about the resulting structure can be gained from the local behaviour of the differential equation. Additionally, computations can be done with computationally convenient Gaussian Markov random fields which approximate the true GRFs. The model is applied to a dataset of annual precipitation in the conterminous US. The non-stationary model performs better than a stationary model measured with both CRPS and the logarithmic scoring rule.

stat.ME↗

arXiv2013

Bayesian computing with INLA: new features

The INLA approach for approximate Bayesian inference for latent Gaussian models has been shown to give fast and accurate estimates of posterior marginals and also to be a valuable tool in practice via the R-package R-INLA. In this paper we formalize new developments in the R-INLA package and show how these features greatly extend the scope of models that can be analyzed by this interface. We also discuss the current default method in R-INLA to approximate posterior marginals of the hyperparameters using only a modest number of evaluations of the joint posterior distribution of the hyperparameters, without any need for numerical integration.

stat.CO↗

arXiv2012

On the convergence of the Escalator Boxcar Train

The Escalator Boxcar Train (EBT) is a numerical method that is widely used in theoretical biology to investigate the dynamics of physiologically structured population models, i.e., models in which individuals differ by size or other physiological characteristics. The method was developed more than two decades ago, but has so far resisted attempts to give a formal proof of convergence. Using a modern framework of measure-valued solutions, we investigate the EBT method and show that the sequence of approximating solution measures generated by the EBT method converges weakly to the true solution measure under weak conditions on the growth rate, birth rate, and mortality rate. In rigorously establishing the convergence of the EBT method, our results pave the way for wider acceptance of the EBT method beyond theoretical biology and constitutes an important step towards integration with established numerical schemes.

math.NA↗

arXiv2012

Bayesian Adaptive Smoothing Spline using Stochastic Differential Equations

The smoothing spline is one of the most popular curve-fitting methods, partly because of empirical evidence supporting its effectiveness and partly because of its elegant mathematical formulation. However, there are two obstacles that restrict the use of smoothing spline in practical statistical work. Firstly, it becomes computationally prohibitive for large data sets because the number of basis functions roughly equals the sample size. Secondly, its global smoothing parameter can only provide constant amount of smoothing, which often results in poor performances when estimating inhomogeneous functions. In this work, we introduce a class of adaptive smoothing spline models that is derived by solving certain stochastic differential equations with finite element methods. The solution extends the smoothing parameter to a continuous data-driven function, which is able to capture the change of the smoothness of underlying process. The new model is Markovian, which makes Bayesian computation fast. A simulation study and real data example are presented to demonstrate the effectiveness of our method.

math.ST↗

arXiv2012

The use of systems of stochastic PDEs as priors for multivariate models with discrete structures

A challenge in multivariate problems with discrete structures is the inclusion of prior information that may differ in each separate structure. A particular example of this is seismic amplitude versus angle (AVA) inversion to elastic parameters, where the discrete structures are geologic layers. Recently, the use of systems of linear stocastic partial differential equations (SPDEs) have become a popular tool for specifying priors in latent Gaussian models. This approach allows for flexible incorporation of nonstationarity and anisotropy in the prior model. Another advantage is that the prior field is Markovian and therefore the precision matrix is very sparse, introducing huge computational and memory benefits. We present a novel approach for parametrising correlations that differ in the different discrete structures, and additionally a geodesic blending approach for quantifying fuzziness of interfaces between the structures. Keywords: Gaussian distribution, multivariate, stochastic PDEs, discrete structures

stat.ME↗

arXiv2011

Think continuous: Markovian Gaussian models in spatial statistics

Gaussian Markov random fields (GMRFs) are frequently used as computationally efficient models in spatial statistics. Unfortunately, it has traditionally been difficult to link GMRFs with the more traditional Gaussian random field models as the Markov property is difficult to deploy in continuous space. Following the pioneering work of Lindgren et al. (2011), we expound on the link between Markovian Gaussian random fields and GMRFs. In particular, we discuss the theoretical and practical aspects of fast computation with continuously specified Markovian Gaussian random fields, as well as the clear advantages they offer in terms of clear, parsimonious and interpretable models of anisotropy and non-stationarity.

math.ST↗

arXiv2011

Fast approximate inference with INLA: the past, the present and the future

Latent Gaussian models are an extremely popular, flexible class of models. Bayesian inference for these models is, however, tricky and time consuming. Recently, Rue, Martino and Chopin introduced the Integrated Nested Laplace Approximation (INLA) method for deterministic fast approximate inference. In this paper, we outline the INLA approximation and its related R package. We will discuss the newer components of the r-INLA program as well as some possible extensions.

stat.CO↗