A Probabilistic Graphical Models Approach to Model Interconnectedness

15 Pages Posted: 30 Nov 2017 Last revised: 16 Mar 2019

See all articles by Alexander Denev

Alexander Denev

Deloitte Financial Advisory Services; University of Oxford

Adrien Papaioannou

IHS Markit

Orazio Angelini

IHS Markit

Date Written: November 27, 2017

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 machine learning technique 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, Stress Testing, Model Risk

Suggested Citation

Denev, Alexander and Papaioannou, Adrien and Angelini, Orazio, A Probabilistic Graphical Models Approach to Model Interconnectedness (November 27, 2017). Available at SSRN: https://ssrn.com/abstract=3078021 or http://dx.doi.org/10.2139/ssrn.3078021

Alexander Denev (Contact Author)

Deloitte Financial Advisory Services ( email )

Amsterdam
Netherlands

University of Oxford ( email )

Mansfield Road
Oxford, Oxfordshire OX1 4AU
United Kingdom

Adrien Papaioannou

IHS Markit ( email )

25 Ropemaker Street
4th floor Ropemaker Place
London, EC2Y 9LY
United Kingdom

Orazio Angelini

IHS Markit ( email )

25 Ropemaker Street
4th floor Ropemaker Place
London, EC2Y 9LY
United Kingdom

Register to save articles to
your library

Register

Paper statistics

Downloads
152
Abstract Views
728
rank
196,312
PlumX Metrics