A1242
Title: Inference and applications of multivariate integer-valued time series
Authors: Konstantinos Fokianos - University of Cyprus (Cyprus) [presenting]
Abstract: Inference and modelling techniques for multivariate count time series are studied, with particular focus on autoregressive, network, and spacetime models. The development of both the underlying theory and its applications depends critically on the choice of an appropriate response distribution for multivariate count data, as well as on the framework of observation-driven models. Recent methodological advances in this area are highlighted and several directions for future research are outlined.