CENTRAL-07 Modelling Dependencies in Ultra-High Dimensions

Modelling dependencies of high-dimensional data plays an important role in economics and statistics. The understanding of these interconnections is essential in actuarial sciences where portfolios of claims are investigated. Our modelling framework will consider various dependence structures and improves previous claim reservations based on assumptions of independency. Further research focus will also be given to systemic risk, extreme events and a copula approach. This partnership will provide the necessary skills and techniques in this area. A major goal of the project is to strengthen the collaboration between the respective research groups in economics and statistics. Cooperation between these institutions over the past few years has already resulted in scientific activities (workshops, seminars and dual learning lectures of statistics), research articles and publications which have been widely used as material for lectures: Härdle, Hlavka (2007): Multivariate Statistics Exercises ; Jaworski, Durante, Härdle (2013): Copulae in Mathematical and Quantitative Finance. Thus, several exchange stays will be arranged with bilateral lectures and seminars as part of this programme. An important milestone of this joint project will also be the expansion of our cooperation in a digital scientific field with the implementation of two digital platforms: Quantnet and MMStat into this DAAD project. The priority of both platforms will be given to transparency and reproducibility. All material (learning notes in MMStat as well as programmed codes in Quantnet) are available for students and researchers even outside of Europe (and throughout the world). MMStat was created in German and English but Polish and Czech versions are also in progress. This cooperation, therefore, will focus on the further development of this platform in these languages as well as including a platform into the lecture and research processes at participating institutions. Cooperation will be strongly supported by the mobility of researchers with regular meetings and participation at international conferences. Workshops (in Vienna, Prague, Warsaw and Berlin) will be organised every year by one of the participating universities. On-going bilateral lectures in statistics will be reinforced further with the main aim of progressing to dual degree programmes in the future.

Härdle, Wolfgang Prof. Dr. (Details) (Statistik)

Projektstart: 01/2015
Projektende: 12/2018

Zuletzt aktualisiert 2021-12-10 um 09:56