Modeling multivariate spatial dependence with copulas: a novel approach

M. Úbeda Flores

Copulas—multivariate distribution functions with univariate marginals that are uniform on the unit interval—offer an innovative alternative to traditional geostatistical methods by modeling spatial and spatio-temporal dependences without the constraint of assuming Gaussian processes. Through Sklar's theorem, copulas allow for the capture of non-Gaussian and asymmetric dependence structures, making them useful in geospatial data analysis. Applications include spatial interpolation, natural process modeling, and climate dependence analysis, demonstrating their flexibility in studying spatial variability.

Palabras clave: Copula spatial dependence geostatistics

E. García Gómez, D. Morales

C. Patino Alonso, M. Gómez Sánchez, L. Gómez Sánchez, S. González Sánchez, C. Agudo Conde

L. Acosta, X. Espuña, J. A. Sanchez-Espigares

D. L. Tarruella Hernández, A. García Molina, J. B. Salom Sanvalero, M. M. Dolcet Negre, M. J. Rivas Lopez

A. García Nogales, P. Perez Fernandez

A. F. Antivilo Bruna, C. Patino Alonso

Á. de Prado Saborido, M. Mirás Calvo, I. Núñez Lugilde, C. Quinteiro Sandomingo, E. Sánchez Rodríguez, A. Bernárdez Ferradás

N. González García, A. B. Nieto Librero

A. Olivares Canal, V. Peña Pizarro, M. Jauch

K. Díaz Arias, I. Albarran Lozano, A. Grané Chávez

P. Gargallo Valero, L. Lample Gracia, J. Miguel Álvarez, M. Salvador Figueras

M. Anciones Polo, A. Queiruga-Dios, P. Vicente Galindo, E. Benéitez-Andrés, M. Rodríguez-Rosa