Title: Time series modeling on dynamic networks
Authors: Jonas Krampe - University of Mannheim (Germany) [presenting]
Abstract: Multivariate time series on dynamic networks with a fixed number of vertices are considered. Each component of the time series is assigned to a vertex of the underlying network. The dynamic dependency between the various components of the time series is modeled by means of the edges. We make use of a multivariate doubly stochastic time series framework, that is we assume linear processes for which the coefficient matrices are stochastic processes themselves. We explicitly allow for dependence in the dynamics of the coefficient matrices, including of course an i.i.d. structure as is typically assumed in random coefficients models. Conditions for stationarity will be given and asymptotic normality of simple statistics like the sample mean is investigated. Furthermore, autoregressive moving average models are defined in this framework. Estimators of the parameters are discussed and how this can be used to forecast such a process. Some interesting features of these processes are shown in simulations and the finite sample behavior of the forecast approach is investigated.