Barclays predictively partners with Simudyne for risk management
Barclays has signed an enterprise-wide licence agreement with London-based simulation company Simudyne to manage risk in a simulated virtual environment.
The simulations that Barclays has been building with Simudyne’s toolkit use a technique known as “agent-based modelling” for quantitative analysis.
C.S. Venkatakrishnan, Barclays’ chief risk officer, says by building agent-based models “we hope to spot and prepare for risks arising from dynamic and large, direct and contingent, counter-party exposures”.
According to the duo, agent-based models allow businesses to simulate complex networks, like financial markets, by examining how the interactions of large individual agents may drive broader behaviour.
Keen to explain their rationale more, they add that unlike traditional “regression-based” models, which analyse the historical behaviour of numerous participants and are based on large-scale historical data analysis, the bottom-up approach of agent-based modelling is used to identify and model risk in instances with sparse historical data and relatively few participants.
Simudyne is also helping Barclays by hosting a series of hands-on workshops and seminars delivered by experts, academics and Simudyne staff. This work will form the basis of a computational simulation centre of excellence inside the bank.
Barclays first began working with Simudyne when it took part in the 2017 cohort of the Barclays Accelerator, powered by Techstars, in London.