Isotropic random fields with infinitely divisible marginal distributions

ABSTRACTA stochastic representation is derived for a vector random field that is stationary, isotropic, and mean square continuous on a sphere or unit circle. The established stochastic representation is an infinite series involving the sequence of ultraspherical or Gegenbauer's polynomials, looks like a mimic of the series representation of the covariance matrix function of the isotropic vector random field, but differs from the spectral representation in terms of the ordinary spherical harmonics. It is also shown in this paper that some isotropic and continuous covariance matrix functions on the real line or , if they are compactly supported, can be adopted as covariance matrix functions on the unit circle or .

This paper is concerned with spherically invariant or elliptically contoured vector random fields in space and/or time, which are formulated as scale mixtures of vector Gaussian random fields. While a spherically invariant vector random field may or may not have second-order moments, a spherically invariant second-order vector random field is determined by its mean and covariance matrix functions, just like the Gaussian one. This paper explores basic properties of spherically invariant second-order vector random fields, and proposes an efficient approach to develop covariance matrix functions for such vector random fields.

The paper tackles the problem of simulating isotropic vector-valued Gaussian random fields defined over the unit two-dimensional sphere embedded in the three-dimensional Euclidean space. Such random fields are used in different disciplines of the natural sciences to model observations located on the Earth or in the sky, or direction-dependent subsoil properties measured along borehole core samples. The simulation is obtained through a weighted sum of finitely many spherical harmonics with random degrees and orders, which allows accurately reproducing the desired multivariate covariance structure, a construction that can actually be generalized to the simulation of isotropic vector random fields on the d-dimensional sphere. The proposed algorithm is illustrated with the simulation of bivariate random fields whose covariances belong to the $${{{\mathcal{F}}}}$$ , spectral Matern and negative binomial classes of covariance functions on the two-dimensional sphere.

Quantitative studies in many fields involve the analysis of multivariate data of diverse types, including measurements that we may consider binary, ordinal and continuous. One approach to the analysis of such mixed data is to use a copula model, in which the associations among the variables are parameterized separately from their univariate marginal distributions. The purpose of this article is to provide a simple, general method of semiparametric inference for copula models via a type of rank likelihood function for the association parameters. The proposed method of inference can be viewed as a generalization of marginal likelihood estimation, in which inference for a parameter of interest is based on a summary statistic whose sampling distribution is not a function of any nuisance parameters. In the context of copula estimation, the extended rank likelihood is a function of the association parameters only and its applicability does not depend on any assumptions about the marginal distributions of the data, thus making it appropriate for the analysis of mixed continuous and discrete data with arbitrary marginal distributions. Estimation and inference for parameters of the Gaussian copula are available via a straightforward Markov chain Monte Carlo algorithm based on Gibbs sampling. Specification of prior distributions or a parametric form for the univariate marginal distributions of the data is not necessary.

Soil nails used to support excavations may be inserted into soils with bedding structures at different orientations. This paper examines for the first time the influence of such cases on probabilistic margins of safety for a typical soil nail arrangement by assigning rotated anisotropic random fields of soil strength parameters that are aligned with the bedding structure of the soil domain. The results using rotated anisotropic random fields on statistical outcomes for the computed global factor of safety are compared to results using homogeneous and isotropic random fields, and deterministic analyses. Analyses using rotated anisotropic random fields show that there is a detectable worst bedding direction that gives the highest probability of failure. However, the range of factor of safety for all random field cases is small and remains in the vicinity of the deterministic value. This is ascribed to the ability of the soil nails to anchor the entire reinforced soil domain together and prevent preferential failure paths to develop through the soil. The paper includes the calculation of the confidence in the point estimate of probability of failure for different conditions, and lessons learned in selecting domain size and level of discretization using the random finite difference method.

First we show that all genetic algorithms can be approximated by an algorithm which keeps the population in linkage equilibrium. Here the genetic population is given as a product of univariate marginal distributions. We describe a simple algorithm which keeps the population in linkage equilibrium. It is called the univariate marginal distribution algorithm (UMDA). Our main result is that UMDA transforms the discrete optimization problem into a continuous one defined by the average fitness W(p1, . . . , p n ) as a function of the univariate marginal distributions p i. For proportionate selection UMDA performs gradient ascent in the landscape defined by W(p). We derive a difference equation for p i which has already been proposed by Wright in population genetics. We show that UMDA solves difficult multimodal optimization problems. For functions with highly correlated variables UMDA has to be extended. The factorized distribution algorithm (FDA) uses a factorization into marginal and conditional distributions. For decomposable functions the optimal factorization can be explicitly computed. In general it has to be computed from the data. This is done by LFDA. It uses a Bayesian network to represent the distribution. Computing the network structure from the data is called learning in Bayesian network theory. The problem of finding a minimal structure which explains the data is discussed in detail. It is shown that the Bayesian information criterion is a good score for this problem.

Needlet approximation for isotropic random fields on the sphere

The estimator of the mean-square approximation of 3-D homogeneous and isotropic random field is investigated. The problem of statistical simulation of realizations of random fields in threedimensional space is considered. The algorithm for the receiving of this realization has been formulated, which has been constructed on the base the mean-square approximation of random fields estimator. It has been constructed the statistical model for the Gaussian random fields in three-dimensional space, which has been given by its statistical characteristics.

This paper investigates spatial data on the unit sphere. Traditionally, isotropic Gaussian random fields are considered as the underlying mathematical model of the cosmic microwave background (CMB) data. We discuss the generalized multifractional Brownian motion and its pointwise Hölder exponent on the sphere. The multifractional approach is used to investigate the CMB data from the Planck mission. These data consist of CMB radiation measurements at narrow angles of the sky sphere. The results obtained suggest that the estimated Hölder exponents for different CMB regions do change from location to location. Therefore, the CMB temperature intensities are multifractional. The methodology developed is used to suggest two approaches for detecting regions with anomalies in the cleaned CMB maps. doi:10.1017/S1446181122000104

This article introduces the hyperbolic vector random field whose finite-dimensional distributions are the generalized hyperbolic one, which is formulated as a scale mixture of Gaussian random fields and is thus an elliptically contoured (or spherically invariant) random field. Such a vector random field may or may not have second-order moments, while a second-order one is characterized by its mean function and its covariance matrix function, just as in a Gaussian case. Some covariance matrix structures of hyperbolic type are constructed in this paper for second-order hyperbolic vector random fields.

This paper gives the exact solution in terms of the Karhunen-Lo\`{e}ve expansion to a fractional stochastic partial differential equation on the unit sphere $\mathbb{S}^{2}\subset \mathbb{R}^{3}$ with fractional Brownian motion as driving noise and with random initial condition given by a fractional stochastic Cauchy problem. A numerical approximation to the solution is given by truncating the Karhunen-Lo\`{e}ve expansion. We show the convergence rates of the truncation errors in degree and the mean square approximation errors in time. Numerical examples using an isotropic Gaussian random field as initial condition and simulations of evolution of cosmic microwave background (CMB) are given to illustrate the theoretical results.

Thousands of permanent GNSS stations sample nowadays the 3D deformation of the Earth’s surface. The temporal covariance structure of the field of GNSS station displacements is well characterized and modelled. On the other hand, there lacks a general agreed-upon model of its spatial covariance structure, in part because the theory of random vector fields on the sphere remains hardly developed.In this contribution, we show how the well-established theory of random isotropic scalar fields on the sphere generalizes to the case of vector fields. We derive in particular a spectral representation of random isotropic vector fields on the sphere in the domain of vector spherical harmonics, from which several properties of their covariance functions follow. We then present several parametric families of covariance functions which could be used to describe and model the covariance structure of vector fields on the sphere, such as GNSS station displacements.Although this presentation focuses on theoretical aspects, it is given with future practical applications in mind. A realistic spatio-temporal covariance model of GNSS station displacements could indeed benefit different problems such as the estimation of long-term GNSS station velocities, the identification and mitigation of offsets in GNSS station position time series, the filtering of spatial “common modes” to isolate local deformation, or the spatial interpolation of GNSS station displacements into global maps. Applications may also be found in other domains involving vector quantities distributed on a sphere, e.g., winds, ocean currents, magnetic anomalies, etc.

We construct matrix-valued covariance functions and in ℝ2 and ℝ3, starting from an arbitrary scalar-valued variogram. It is shown that sufficiently smooth random vector fields (RVFs) with these covariance functions have divergence-free and curl-free sample paths, respectively. Conversely, essentially all models with such sample paths can be obtained via our construction. Extensions to space-time RVFs are possible. RVFs with divergence-free and curl-free sample paths can be utilised in meteorological applications e.g. for modelling and interpolating wind fields.

Suppose k-variate data are drawn from a mixture of two distributions, each having independent components. It is desired to estimate the univariate marginal distributions in each of the products, as well as the mixing proportion. This is the setting of two-class, fully parametrized latent models that has been proposed for estimating the distributions of medical test results when disease status is unavailable. The problem is one of inference in a mixture of distributions without training data, and until now it has been tackled only in a fully parametric setting. We investigate the possibility of using nonparametric methods. Of course, when k=1 the problem is not identifiable from a nonparametric viewpoint. We show that the problem is "almost" identifiable when k=2; there, the set of all possible representations can be expressed, in terms of any one of those representations, as a two-parameter family. Furthermore, it is proved that when $k\geq3$ the problem is nonparametrically identifiable under particularly mild regularity conditions. In this case we introduce root-n consistent nonparametric estimators of the 2k univariate marginal distributions and the mixing proportion. Finite-sample and asymptotic properties of the estimators are described.

Best Attainable Rates of Convergence for Estimators of the Stable Tail Dependence Function