Title: A probabilistic graphical models approach to model interconnectedness

Authors: Alexander Denev; Adrien Papaioannou; Orazio Angelini

Addresses: University of Oxford, Wellington Square, Oxford, OX1 2JD, UK; IHS Markit, 4th floor Ropemaker Place, 25 Ropemaker Street, London EC2Y 9LY, UK ' University of Oxford, Wellington Square, Oxford, OX1 2JD, UK; IHS Markit, 4th floor Ropemaker Place, 25 Ropemaker Street, London EC2Y 9LY, UK ' King's College, Strand, WC2R 2LS, London, UK

Abstract: In this paper, we show that using multiple models when executing a specific task almost unavoidably gives rise to interaction between them, especially when their number is large. We show that this interaction can lead to biased and incomplete results if treated inappropriately (which we believe is the current standard in the financial industry). We propose the use of probabilistic graphical models – a technique widely used in machine learning and expert systems as a remedy to this problem. We discuss some numerical aspects of our approach that will be present in any practical implementation. We then examine, in detail, a practical example of using this method in a stress testing context.

Keywords: probabilistic graphical models; model interconnectedness; stress testing; machine learning.

DOI: 10.1504/IJRAM.2020.106963

International Journal of Risk Assessment and Management, 2020 Vol.23 No.2, pp.119 - 133

Received: 07 Apr 2018
Accepted: 31 Dec 2018
Published online: 29 Apr 2020 *

Full-text access for editors Full-text access for subscribers Purchase this article Comment on this article