One needs to make use of the sharing tools that are found via the share button at the top or side of articles. Copying articles in order to share with others is a breach of FT.com T&Cs as well as Copyright Policy. One has to email in order to buy additional rights. Subscribers may indeed share up to 10 or 20 articles per month making use of the gift article service.
Artificial intelligence can indeed soon be powering a new way of making decisions across Barclays’ trading; lending as well as risk management divisions after the UK bank became the flagship customer of AI simulation firm Simudyne Technology. Simudyne does allow banks to create computer models that do simulate millions of possible future scenarios, thus allowing them to test how individual factors will indeed perform as well as interact with each other in a vast array of situations.
As of now banks largely do rely on models that extrapolate data from historical data, complemented by complex as well as expensive simulations in niche areas such as valuing options. Recommended Simulations are rather based on agent-based modeling techniques which do allow banks to account for the ability of agents — the people, firms, traders and so on in order to deviate from rationality, optimize or exploit their respective environment.
Given a policy change, these models can indeed be used to see how a system will reconfigure itself thus allowing machine intelligence to stay ahead of the human responses. Focus has been on exploring “tail events” such as trade wars or other unpredictable policy outcomes. Efforts have been on to highlight the potential for agent-based models to address complex macroeconomic questions, particularly those whereby heterogeneity, networks, as well as heuristics that do play an important role. The bank did hope agent-based models would rather help it to spot as well as prepare for risks arising from dynamic as well as large, direct and contingent counter-party exposures.
The most elaborate modelling efforts to date have been much focused on their trading rooms. The software does enable them to recreate the factors that are underlying an entire financial marketplace. The most interesting things that can actually happen — the collapse of a bank, the spread of contagion — are the result of multiple factors that aggregate towards a tipping point.
For lending, the software that allows banks to build simulations that is based on inputs such as household incomes, how people do spend their money and how they also behave as borrowers. An example is that a team at Barclays is exploring the model of the actions of a fraudster.
Artificial intelligence is gaining ground these days and even banks are seeking such measures in order to carry out their banking processes. It is a competitive world and it is necessary to streamline procedures and this applies to banking as well. Nowadays, banks also opt for technology that would expedite their official matters and make matters easy to operate for them.
