Incorporating Spatial Interaction between Large Dimensional Temperature Series in Atmosphere –Ocean Modelling of Global Climate Change

Functional Statistics provides a suitable framework for the analysis of large dimensional data sets. In this paper, we consider the spatial autoregressive functional series (SARH(1)) framework. This framework allows the incorporation of spatial interaction to the statistical analysis of functional data (see Ruiz-Medina [1], , [2]). The SARH(1) model is fitted to the average ocean temporal temperature profiles, collected in different spatial locations, during the years 1998, 1999 and 2000. We have used the public oceanographic bio-optical database, The World-wide Ocean Optics Database (WOOD). The SARH(1) plug-in extrapolator is then computed, from the previously derived infinite-dimensional projection parameter estimator.