Detailansicht

Poisson count processes: Mixing properties and statistical methods | Prof. Michael Neumann (Universität Jena)

Kurzbeschreibung:
Startdatum: 24.11.2017 - 16:00
Enddatum: 24.11.2017 - 17:30
Adresse: MZH 6210
Organisator/Ansprechpartner:,
Preis: 0€

Models for integer-valued time series occur in many areas, ranging from statistics and econometrics to social and physical sciences.
A popular class is given by processes where the observable random variables conditioned on the past have a Poisson distribution and the current value of the accompanying intensity process depends on previous values of both processes. We show stationarity and mixing properties of these processes and apply these results to an asymptotic justification of statistical methods.

Einladung von Prof. Thorsten Dickhaus (Institut für Statistik)