CFE 2020: Start Registration
View Submission - CMStatistics
B0278
Title: The multinomial micro-level reserving model Authors:  Emmanuel Jordy Menvouta - KU Leuven (Belgium)
Jolien Ponnet - KU Leuven (Belgium)
Tim Verdonck - UAntwerp, KU Leuven (Belgium)
Christopher Grumiau - Allianz Benelux (Belgium)
Robin Van Oirbeek - UAntwerp (Belgium) [presenting]
Abstract: The estimation of the open liabilities or claims reserve is a very important exercise to ensure the day-to-day activities and even the financial viability of an non-life insurance company. Typically, macro-level reserving models such as the Chain Ladder, are used to this end, estimating the claims reserve for the entire portfolio. However, it will be shown how the claims reserve can be estimated on claim-by-claim basis by the use of a micro-level reserving model. The latter models capture the entire lifecycle of a claim by explicitly modelling the underlying time and payment process. Specific to this version of the micro-level reserving model is that both processes, as well as the IBNR or 'Incurred But Not Reported' model are all modeled using separate multinomial regression models. How all these models are tied together, will be discussed.