engleski

Record times of stationary regularly varying time series

In this talk we present a new type of the so{; ; called complete point process convergence result for a stationary, regularly varying and weakly dependent time series. Due to dependence, extremal observations of appear in clusters and the main novelty is that this convergence preserves the information about the temporal ordering of observations belonging to the same cluster. The key tool is the so-called tail process of the corresponding time series. As an application, we deduce the asymptotic behavior of rescaled record times, extending the well-known result in the i.i.d. case. These results are based on the joint work with Bojan Basrak and Philippe Soulier.

Point process ; Tail process ; Record times

nije evidentirano