GAMs and functional kriging for air quality data

Abstract

Data having spatio-temporal structure are often observed in environmental sciences. They may be considered as discrete observations from curves along time and/or space and treated as functional. Generalized Additive Models (GAMs) represent a useful tool for modelling, for example, as pollutant concentrations describing their spatial and/or temporal trends.Usually, the prediction of a curve at an unmonitored site is necessary and, with this aim, we extend kriging for functional data to a multivariate context. Moreover, even if we are interested only in predicting a single pollutant, such as PM10, the estimation can be improved exploiting its correlation with the other pollutants. Cross validation is used to test the performance of the proposed procedure.