A note on defining positive definite functions
Authors:
Lucas da Cunha Godoy,
Marcos Oliveira Prates,
Fernando Andrés Quintana
Abstract:
A fundamental requirement in spatial statistics is that covariance functions are positive definite. While many positive definite functions are known for Euclidean spaces, their positive definiteness may not extend to non-Euclidean spaces. We present sufficient conditions to derive valid positive definite functions for spatial statistics on non-Euclidean geometries. Our approach leverages overlooke…
▽ More
A fundamental requirement in spatial statistics is that covariance functions are positive definite. While many positive definite functions are known for Euclidean spaces, their positive definiteness may not extend to non-Euclidean spaces. We present sufficient conditions to derive valid positive definite functions for spatial statistics on non-Euclidean geometries. Our approach leverages overlooked results due to Schoenberg (1938) to establish conditions under which covariance functions that are valid on Euclidean spaces remain valid on the domain of interest. This approach provides a more accessible and direct framework for spatial statisticians with diverse backgrounds.
△ Less
Submitted 20 February, 2025;
originally announced February 2025.
Statistical Inferences and Predictions for Areal Data and Spatial Data Fusion with Hausdorff--Gaussian Processes
Authors:
Lucas da Cunha Godoy,
Marcos Oliveira Prates,
Jun Yan
Abstract:
Accurate modeling of spatial dependence is pivotal in analyzing spatial data, influencing parameter estimation and predictions. The spatial structure of the data significantly impacts valid statistical inference. Existing models for areal data often rely on adjacency matrices, struggling to differentiate between polygons of varying sizes and shapes. Conversely, data fusion models rely on computati…
▽ More
Accurate modeling of spatial dependence is pivotal in analyzing spatial data, influencing parameter estimation and predictions. The spatial structure of the data significantly impacts valid statistical inference. Existing models for areal data often rely on adjacency matrices, struggling to differentiate between polygons of varying sizes and shapes. Conversely, data fusion models rely on computationally intensive numerical integrals, presenting challenges for moderately large datasets. In response to these issues, we propose the Hausdorff-Gaussian process (HGP), a versatile model utilizing the Hausdorff distance to capture spatial dependence in both point and areal data. Integration into generalized linear mixed-effects models enhances its applicability, particularly in addressing data fusion challenges. We validate our approach through a comprehensive simulation study and application to two real-world scenarios: one involving areal data and another demonstrating its effectiveness in data fusion. The results suggest that the HGP is competitive with specialized models regarding goodness-of-fit and prediction performances. In summary, the HGP offers a flexible and robust solution for modeling spatial data of various types and shapes, with potential applications spanning fields such as public health and climate science.
△ Less
Submitted 21 February, 2025; v1 submitted 16 August, 2022;
originally announced August 2022.
