Moreno Bevilacqua

Estimating Space and Space-Time Covariance Functions for Large Data Sets: A Weighted Composite Likelihood Approach

In this article, we propose two methods for estimating space and space-time covariance functions from a Gaussian random field, based on the composite likelihood idea. The first method relies on the maximization of a weighted version of the composite likelihood function, while the second one is based on the solution of a weighted composite score equation. This last scheme is quite general and could be applied to any kind of composite likelihood. An information criterion for model selection based on the first estimation method is also introduced. The methods are useful for practitioners looking for a good balance between computational complexity and statistical efficiency. The effectiveness of the methods is illustrated through examples, simulation experiments, and by analyzing a dataset on ozone measurements.

Spatio-Temporal Covariance and Cross-Covariance Functions of the Great Circle Distance on a Sphere

Abstract In this article, we propose stationary covariance functions for processes that evolve temporally over a sphere, as well as cross-covariance functions for multivariate random fields defined over a sphere. For such processes, the great circle distance is the natural metric that should be used to describe spatial dependence. Given the mathematical difficulties for the construction of covariance functions for processes defined over spheres cross time, approximations of the state of nature have been proposed in the literature by using the Euclidean (based on map projections) and the chordal distances. We present several methods of construction based on the great circle distance and provide closed-form expressions for both spatio-temporal and multivariate cases. A simulation study assesses the discrepancy between the great circle distance, chordal distance, and Euclidean distance based on a map projection both in terms of estimation and prediction in a space-time and a bivariate spatial setting, where the space is in this case the Earth. We revisit the analysis of Total Ozone Mapping Spectrometer (TOMS) data and investigate differences in terms of estimation and prediction between the aforementioned distance-based approaches. Both simulation and real data highlight sensible differences in terms of estimation of the spatial scale parameter. As far as prediction is concerned, the differences can be appreciated only when the interpoint distances are large, as demonstrated by an illustrative example. Supplementary materials for this article are available online.

Comparing composite likelihood methods based on pairs for spatial Gaussian random fields

In the last years there has been a growing interest in proposing methods for estimating covariance functions for geostatistical data. Among these, maximum likelihood estimators have nice features when we deal with a Gaussian model. However maximum likelihood becomes impractical when the number of observations is very large. In this work we review some solutions and we contrast them in terms of loss of statistical efficiency and computational burden. Specifically we focus on three types of weighted composite likelihood functions based on pairs and we compare them with the method of covariance tapering. Asymptotic properties of the three estimation methods are derived. We illustrate the effectiveness of the methods through theoretical examples, simulation experiments and by analyzing a data set on yearly total precipitation anomalies at weather stations in the United States.

Radial basis functions with compact support for multivariate geostatistics

Matrix-valued radially symmetric covariance functions (also called radial basis functions in the numerical analysis literature) are crucial for the analysis, inference and prediction of Gaussian vector-valued random fields. This paper provides different methodologies for the construction of matrix-valued mappings that are positive definite and compactly supported over the sphere of a d-dimensional space, of a given radius. In particular, we offer a representation based on scaled mixtures of Askey functions; we also suggest a method of construction based on B-splines. Finally, we show that the very appealing convolution arguments are indeed effective when working in one dimension, prohibitive in two and feasible, but substantially useless, when working in three dimensions. We exhibit the statistical performance of the proposed models through simulation study and then discuss the computational gains that come from our constructions when the parameters are estimated via maximum likelihood. We finally apply our constructions to a North American Pacific Northwest temperatures dataset.

Classes of compactly supported covariance functions for multivariate random fields

The paper combines simple general methodologies to obtain new classes of matrix-valued covariance functions that have two important properties: (i) the domains of the compact support of the several components of the matrix-valued functions can vary between components; and (ii) the overall differentiability at the origin can also vary. These models exploit a class of functions called here the Wendland–Gneiting class; their use is illustrated via both a simulation study and an application to a North American bivariate dataset of precipitation and temperature. Because for this dataset, as for others, the empirical covariances exhibit a hole effect, the turning bands operator is extended to matrix-valued covariance functions so as to obtain matrix-valued covariance models with negative covariances.

Modelling residuals dependence in dynamic life tables: A geostatistical approach

The problem of modelling dynamic mortality tables is considered. In this context, the influence of age on data graduation needs to be properly assessed through a dynamic model, as mortality progresses over the years. After detrending the raw data, the residuals dependence structure is analysed, by considering them as a realisation of a homogeneous Gaussian random field defined on RxR. This setting allows for the implementation of geostatistical techniques for the estimation of the dependence and further interpolation in the domain of interest. In particular, a complex form of interaction between age and time is considered, by taking into account a zonally anisotropic component embedded into a nonseparable covariance structure. The estimated structure is then used for prediction of mortality rates, and goodness-of-fit testing is performed through some cross-validation techniques. Comments on validity and interpretation of the results are given.

Weighted composite likelihood-based tests for space-time separability of covariance functions

Testing for separability of space-time covariance functions is of great interest in the analysis of space-time data. In this paper we work in a parametric framework and consider the case when the parameter identifying the case of separability of the associated space-time covariance lies on the boundary of the parametric space. This situation is frequently encountered in space-time geostatistics. It is known that classical methods such as likelihood ratio test may fail in this case.We present two tests based on weighted composite likelihood estimates and the bootstrap method, and evaluate their performance through an extensive simulation study as well as an application to Irish wind speeds. The tests are performed with respect to a new class of covariance functions, which presents some desirable mathematical features and has margins of the Generalized Cauchy type. We also apply the test on a element of the Gneiting class, obtaining concordant results.

Covariance tapering for multivariate Gaussian random fields estimation

In recent literature there has been a growing interest in the construction of covariance models for multivariate Gaussian random fields. However, effective estimation methods for these models are somehow unexplored. The maximum likelihood method has attractive features, but when we deal with large data sets this solution becomes impractical, so computationally efficient solutions have to be devised. In this paper we explore the use of the covariance tapering method for the estimation of multivariate covariance models. In particular, through a simulation study, we compare the use of simple separable tapers with more flexible multivariate tapers recently proposed in the literature and we discuss the asymptotic properties of the method under increasing domain asymptotics.

Validity of covariance models for the analysis of geographical variation

Summary Due to the availability of large molecular data sets, covariance models are increasingly used to describe the structure of genetic variation as an alternative to more heavily parameterized biological models. We focus here on a class of parametric covariance models that received sustained attention lately and show that the conditions under which they are valid mathematical models have been overlooked so far. We provide rigorous results for the construction of valid covariance models in this family. We also outline how to construct alternative covariance models for the analysis of geographical variation that are both mathematically well behaved and easily implementable. The full R code to reproduce the numerical analysis is available from: http://www2.imm.dtu.dk/~gigu/MEE_GSPB/covariance_validity.html

Combining Euclidean and composite likelihood for binary spatial data estimation

In this paper we propose a blockwise Euclidean likelihood method for the estimation of a spatial binary field obtained by thresholding a latent Gaussian random field. The moment conditions used in the Euclidean likelihood estimator derive from the score of the composite likelihood based on marginal pairs. A feature of this approach is that it is possible to obtain computational benefits with respect to the pairwise likelihood depending on the choice of the spatial blocks. A simulation study and an analysis on cancer mortality data compares the two methods in terms of statistical and computational efficiency. We also study the asymptotic properties of the proposed estimator.