152
APPLICATION AREAS
•
Financial and Business Services
•
Electronics and electrical systems
•
Information and communication
technology (ICT)
Keywords
Game theory, automated trading, multi-agent systems,
financial modelling
Expertise
Banks rely on the availability of cash liquidity to meet
their obligations to clients and to engage in financial
transactions. The problem is that intraday cash flows,
both within and between banks, are complex and can
be affected by behavioural factors such as fear of
counterparty failure, opportunism, greed, and malice.
Since liquidity crises have an adverse effect on the real
economy, as illustrated during late 2008, it is extremely
important to develop an understanding of liquidity flows and
to develop tools and techniques to monitor cash liquidity
and help to reduce the incidence of liquidity problems.
Although economic models are widely used by banks,
regulators and governments in an attempt to understand
how markets behave, they can exhibit limitations such
as their inability to predict robustly how individual factors
affect the behaviour of the market.
Agent-based models capture such factors explicitly and
transparently, thus offering great potential for understand-
ing the intraday cash liquidity market. They can be used
to directly represent the key players in a market, explicitly
capturing their beliefs, policies and behaviours. The
structure of these models reflects reality, so understanding
its properties is easier than with equation-based models.
As such, agent-based modelling can become a powerful
tool to understand emergent market-based behaviour.
Liverpool has a wealth of expertise in agent-based
modelling, game theory, optimisation, algorithms
and complexity.
Capabilities and facilities
Our focus lies on distributed, multi-party financial
systems; typically we analyse and model disparate,
rational, autonomous entities. Our studies are
underpinned by the exploration and theoretical
characterisation of computational foundations of
economics and game theory and the application
of economic theory in computer science.
Research areas include:
•
Algorithmic game theory
•
Mechanism design and auction theory
•
Complexity and computation of solution
concepts
•
Optimisation problems in economics and
computational social choice.
Our expertise in agent-based modelling of financial
institutions can be used across the financial and business
sector to understand, predict and react to the dynamics
of market behaviour. It can identify triggers for action in
terms of early warning signals and intervention activities,
predicting catastrophic failures due to unforeseen
combinations of events. Our models can also ascertain
the most important structural and behavioural aspects of
current financial systems.
3.
Complex software
3.1
Computational markets and agent-based modelling
Also see:
Risk, Safety & Security –
1.3
Systemic risk in economic and
financial systems, page 117
Management & Enterprise –
3.
Accountability, regulation and
governance, page 161
7.
Finance and economics, page 164
For further information
on all our specialist
centres, facilities and
laboratories
go to page
179
Digital Technologies
