Inferencia bayesiana aproximada para el modelo multivariado block-NNGP
No Thumbnail Available
Date
2025-01-21
Authors
Journal Title
Journal ISSN
Volume Title
Publisher
Pontificia Universidad Católica del Perú
Abstract
El estudio de las especies de aves es un excelente indicador de la biodiversidad o la
productividad. Se sabe que el calentamiento global y los cambios en el uso de la tierra por
parte de los humanos están afectando la abundancia de aves. En este estudio nos enfocamos
en las especies Morning Dove y American Robin, las especies más abundantes en América del
norte. Las abundancias de estas especies pueden estar correlacionadas entre sí y mostrar una
distribución espacial similar. Por lo tanto, proponemos modelar estos datos simultáneamente
a través de modelos multivariados espaciales que se basan en compartir términos comunes de
efectos aleatorios espaciales gaussianos. Para mejorar la eficiencia computacional, los procesos
espaciales gaussianos se aproximan a un proceso gaussiano de vecinos más cercanos por
bloques (block-NNGP). El modelo geoestadístico multivariado pertenece a la clase de modelos
gaussianos latentes, por ello se usó el método de aproximación de Laplace anidada integrada
(INLA) que permite una inferencia bayesiana rápida. El rendimiento del modelo propuesto
se demuestra a través de simulaciones y la aplicación a los datos de especies de aves.
The study of birds species is an excellent indicator of biodiversity or productivity. Global warming and changes human land us are considered major threats to biodiversity, affecting the abundance of bird species. In this study we focus on the Mourning Dove and American Robin, the most abundant birds species in the United States. The abundances of these species can be correlated between them and they would also be similar in nearby locations. Thus we propose to model these data simultaneously through multivariate models that relies on sharing common spatial Gaussian random effect terms. In order to improve the computational efficiency, each spatial Gaussian process is approximated to the block nearest neighbor Gaussian process (block-NNGP). Since the multivariate geostatistical model belongs to the class of Latent Gaussian Models, fast Bayesian inference can be carried out through the Integrated Nested Laplace Approximation (INLA) method. The good performance of the proposed model is shown through simulations and our application to the bird species real data.
The study of birds species is an excellent indicator of biodiversity or productivity. Global warming and changes human land us are considered major threats to biodiversity, affecting the abundance of bird species. In this study we focus on the Mourning Dove and American Robin, the most abundant birds species in the United States. The abundances of these species can be correlated between them and they would also be similar in nearby locations. Thus we propose to model these data simultaneously through multivariate models that relies on sharing common spatial Gaussian random effect terms. In order to improve the computational efficiency, each spatial Gaussian process is approximated to the block nearest neighbor Gaussian process (block-NNGP). Since the multivariate geostatistical model belongs to the class of Latent Gaussian Models, fast Bayesian inference can be carried out through the Integrated Nested Laplace Approximation (INLA) method. The good performance of the proposed model is shown through simulations and our application to the bird species real data.
Description
Keywords
Geología--Métodos estadísticos, Análisis multivariante--Procesamiento de datos, Procesos de Gauss, Aves--América del Norte
Citation
Collections
Endorsement
Review
Supplemented By
Referenced By
Creative Commons license
Except where otherwised noted, this item's license is described as info:eu-repo/semantics/embargoedAccess