A wavelet-based kernel estimator for spatiotemporal prediction

Autores: Rosaura Fernández Pascual , María Dolores Ruiz Medina , José Miguel Angulo Ibáñez
Localización: XXX Congreso Nacional de Estadística e Investigación Operativa y de las IV Jornadas de Estadística Pública: actas, 2007, ISBN 978-84-690-7249-3
Idioma: inglés
Texto completo no disponible (Saber más ...)
Resumen
- Anomalous di®usions usually arise in the description of geophysical and environmental processes. The high local singularity of the physical law governing such processes hinders the resolution of the associated prediction and ¯ltering problems. Speci¯cally, such problems are ill-posed regarding the inversion of the involved Wiener-Hopf equation. In this paper, a kernel estimator based on wavelets is proposed. The local regularity properties and moment conditions of the wavelet basis selected allow the stable inversion of such an equation. The results obtained are compared with those derived when the L2¡geometry is transformed in terms of the reproducing kernel Hilbert-Schmidt geometry to get the continuous inversion (see Ruiz-Medina, Angulo and Fern¶andez-Pascual, 2007).