Ramón Giraldo

A universal kriging approach for spatial functional data

In a wide range of scientific fields the outputs coming from certain measurements often come in form of curves. In this paper we give a solution to the problem of spatial prediction of non-stationary functional data. We propose a new predictor by extending the classical universal kriging predictor for univariate data to the context of functional data. Using an approach similar to that used in univariate geostatistics we obtain a matrix system for estimating the weights of each functional variable on the prediction. The proposed methodology is validated by analyzing a real dataset corresponding to temperature curves obtained in several weather stations of Canada.

Multivariate functional random fields: prediction and optimal sampling

This paper develops spatial prediction of a functional variable at unsampled sites, using functional covariates, that is, we present a functional cokriging method. We show that through the representation of each function in terms of its empirical functional principal components, the functional cokriging only depends on the auto-covariance and cross-covariance of the associated scores vectors, which are scalar random fields. In addition, we propose the methodology to find optimal sampling designs in this context. The proposal is applied to the network of air quality in México city.

Optimal sampling for spatial prediction of functional data

This paper combines optimal spatial sampling designs with geostatistical analysis of functional data. We propose a methodology and design criteria to find the set of spatial locations that minimizes the variance of the spatial functional prediction at unsampled sites for three functional predictors: ordinary kriging, simple kriging and simple cokriging. The last one is a modification of an existing predictor that uses ordinary cokriging based on the basis coefficients. Instead, we propose to use a simple cokriging predictor with the scores resulting from a representation of the functional data with the empirical functional principal components, allowing to remove restrictions and complexity of the covariance models and constraints on the estimation procedure. The methodology is applied to a network of air quality in Bogotá city, Colombia.

Spatial prediction for infinite-dimensional compositional data

There is a growing interest in the analysis of geostatistical functional data. Such a fusion between geostatistical methods and functional data analysis has been shown to open promising area of research. The present paper is devoted to a kriging predictor for functional data where the functions are probability densities functions (PDFs for short), being also a special case of infinite dimensional compositional data. The predictor proposed in this paper is the analogue of the classic ordinary kriging predictor, defined in terms of scalar parameters, but considering PDFs (with support on a finite interval) instead of one-dimensional data. The methodology is applied to both simulated and real data. The statistical performance of our predictor is then evaluated through cross validation techniques using real and simulated data. From the mathematical point of view, in order to assess the properties of our predictor we need to work under the framework of Hilbert valued random fields as well as with Aitchison geometries. We present some original results being of interest for themselves, and that give a complete picture of the framework illustrated through the paper.

Residual Kriging for Functional Spatial Prediction of Salinity Curves

Recently, several methodologies to perform geostatistical analysis of functional data have been proposed. All of them assume that the spatial functional process considered is stationary. However, in practice, we often have nonstationary functional data because there exists an explicit spatial trend in the mean. Here, we propose a methodology to extend kriging predictors for functional data to the case where the mean function is not constant through the region of interest. We consider an approach based on the classical residual kriging method used in univariate geostatistics. We propose a three steps procedure. Initially, a functional regression model is used to detrend the mean. Then we apply kriging methods for functional data to the regression residuals to predict a residual curve at a non-data location. Finally, the prediction curve is obtained as the sum of the trend and the residual prediction. We apply the methodology to salinity data corresponding to 21 salinity curves recorded at the Ciénaga Grande de Santa Marta estuary, located in the Caribbean coast of Colombia. A cross-validation analysis was carried out to track the performance of the proposed methodology.

Clustering Spatially Correlated Functional Data

In this paper we discuss and compare two clustering strategies: a hierarchical clustering and a dynamic clustering method for spatially correlated functional data. Both the approaches aim to obtain clusters which are internally homogeneous in terms of their spatial correlation structure. With this scope they incorporate the spatial information into the clustering process by considering, in a different manner, a measure of spatial association ables to emphasize the average spatial dependence among curves: the trace-variogram function.

Spatial prediction of soil penetration resistance using functional geostatistics

Knowledge of agricultural soils is a relevant factor for the sustainable development of farming activities. Studies on agricultural soils usually begin with the analysis of data obtained from sampling a finite number of sites in a particular region of interest. The variables measured at each site can be scalar (chemical properties) or functional (infiltration water or penetration resistance). The use of functional geostatistics (FG) allows to perform spatial curve interpolation to generate prediction curves (instead of single variables) at sites that lack information. This study analyzed soil penetration resistance (PR) data measured between 0 and 35 cm depth at 75 sites within a 37 ha plot dedicated to livestock. The data from each site were converted to curves using non-parametric smoothing techniques. In this study, a B-splines basis of 18 functions was used to estimate PR curves for each of the 75 sites. The applicability of FG as a spatial prediction tool for PR curves was then evaluated using cross-validation, and the results were compared with classical spatial prediction methods (univariate geostatistics) that are generally used for studying this type of information. We concluded that FG is a reliable tool for analyzing PR because a high correlation was obtained between the observed and predicted curves (R2 = 94 %). In addition, the results from descriptive analyses calculated from field data and FG models were similar for the observed and predicted values.

Inference in log-alpha-power and log-skew-normal multivariate models

ABSTRACT Random vectors with positive components are common in many applied fields, for example, in meteorology, when daily precipitation is measured through a region Marchenko and Genton (2010). Frequently, the log-normal multivariate distribution is used for modeling this type of data. This modeling approach is not appropriate for data with high asymmetry or kurtosis. Consequently, more flexible multivariate distributions than the log-normal multivariate are required. As an alternative to this distribution, we propose the log-alpha-power multivariate and log-skew-normal multivariate models. The first model is an extension for positive data of the fractional order statistics model Durrans (1992). The second one is an extension of the log-skew-normal model studied by Mateu-Figueras and Pawlowsky-Glahn (2007). We study parameter estimation for these models by means of pseudo-likelihood and maximum likelihood methods. We illustrate the proposal analyzing a real dataset.