WG Risk :“Breaking the Markov Barrier in Life Insurance"

The Working Group on Risk - CREAR, with the support of the ESSEC IDO dpt/Ceressec, Institut des actuaires, Labex MME-DII (CY), and the group Risques AEF - SFdS, has the pleasure to invite you to the seminar by :

Prof. Marcus Christiansen, Université Carl von Ossietzky d'Oldenbourg, Germany. 

Dr. Marcus Christiansen is Professor of Applied Probability Theory at the Institute of Mathematics at the Carl von Ossietzky University in Oldenburg, Germany. Previously, he worked at Heriot-Watt University in Edinburgh, the University of Ulm and the Karlsruhe Institute of Technology. Marcus had extended research stays at the Université Catholique de Louvain and the University of Copenhagen. His primary research interest lies in the field of stochastic processes in insurance, particularly life insurance. He is an editor at
the Annals of Actuarial Science and recently became vice-chairman of the German Society for Actuarial and Financial Mathematics (DGVFM).

“Breaking the Markov Barrier in Life Insurance”

The classical Thiele equation is the precursor of the Feynman-Kac equation and describes the time dynamics of conditional expectations with respect to Markovian jump processes. It is a key tool in life insurance not only for numerical calculations, but also for sensitivity analyses, safe-side calculations, surplus calculations and contract modifications. In 1992, Ragnar Norberg generalised the Thiele equation to non-Markovian modelling by introducing a so-called stochastic Thiele equation. This backward stochastic differential equation makes it possible to generalise the common risk management techniques beyond the limited world of Markov modelling, so it is surprising that this generalisation has not yet found its way into life insurance textbooks. Indeed, there have been a number of unresolved technical issues with the stochastic Thiele equation, and we show how they can be overcome. As a result, we are finally able to present the basic concepts of risk management in life insurance without the usual restrictive Markov assumptions.

Dual format:

ESSEC Paris La Défense (CNIT), Room 208, and via Zoom, please click here .