Spectral-marginal-based estimation of spatiotemporal long-range dependence

  1. Frías Bustamante, María del Pilar
  2. Ruiz Medina, María Dolores
  3. Alonso, Francisco J.
  4. Angulo Ibáñez, José Miguel
Libro:
XXX Congreso Nacional de Estadística e Investigación Operativa y de las IV Jornadas de Estadística Pública: actas

Editorial: Comité organizador del XXX Congreso Nacional de Estadística e Investigación Operativa y IV Jornadas de Estadística Pública

ISBN: 978-84-690-7249-3

Año de publicación: 2007

Congreso: Congreso Nacional de Estadística e Investigación Operativa (30. 2007. Valladolid)

Tipo: Aportación congreso

Resumen

Long-memory and strong spatial dependence are two features which can arise jointly or separately depending on the tail behavior of the temporal and spatial covariance functions of a given spatiotemporal process. Under certain conditions, such a behavior can be related to the behavior of temporal and spatial frequencies in a neighborhood of the origin. In particular, a spatiotemporal process displaying long-memory and/or strong spatial dependence can be built, in terms of separable heavy-tail filters, from a second-order processes satisfying suitable regularity and moment conditions. A parameter estimation method based on the marginal spectral densities is implemented to approximate the long-memory and/or strong-spatial-dependence parameters.