B0963
Title: Fast estimation of adaptive P-spline models
Authors: Maria Durban - Universidad Carlos II de Madrid (Spain) [presenting]
Maria Xose Rodriguez-Alvarez - BCAM - Basque Center for Applied Mathematics (Spain)
Dae-Jin Lee - BCAM - Basque Center for Applied Mathematics (Spain)
Paul Eilers - Erasmus University Medical Centre (Netherlands)
Abstract: In many applications it is desirable and needed to adapt smoothness locally to the data, and adaptive P-splines have been suggested. However, the existing estimation procedures can be very slow or even unstable. We extend a previous method, and generalize the Separation of Anisotropic Penalties (SAP) algorithm to deal with the proposed adaptive penalty in one or more dimensions. The practical performance of the algorithm is evaluated by means of simulations, and comparisons with alternative methods are made on the basis of the mean square error criterion and the computing time. Finally, we illustrate our proposal with the analysis of two datasets: one corresponds to photon counts of diffracted x-ray radiation as a function of the angle of diffraction, and the other in which we consider a spatio-temporal adaptive penalized for modelling the firing rate of visual neurons.
Journal