Estimation of the dynamic coregionalization model and air quality monitoring by integrated ground level and satellite data
Lunedý 15 novembre 2010, ore 11.00, Aula 201 - Alessandro FASSO' - UniversitÓ degli Studi di Bergamo -
Introduce: Giuseppe Cicchitelli, UniversitÓ di Perugia
In this paper, we discuss the dynamic coregionalization model and derive new formulas for the EM algorithm which allows to cover maximum likelihood estimation for large data sets with structural missing data.
Moreover we discuss its use in air quality monitoring, namely spatio-temporal dynamic calibration of satellite data and mapping daily concentration of airborne particulate matter. To do this, we consider the problem of joint modelling ground level concentration data and satellite measurements of aerosol optical thickness (AOT) whose availability is related to weather conditions.
Top of Page