Information Technology Reference
In-Depth Information
Fig. 1.
Double auction market with human and artificial traders
sellers and buyers based on an asynchronous, double auction mechanism structured in
an order topic (see Figure 1). ATOM is developed as large scale experimental platform
with heterogeneous agents populations. It is a concrete implementation of MAS ab-
stract design issues:
agent autonomy
and its
behaviour
(strategy), each agent asses her
position and makes decisions individually; price history and orders sets are
emergent
phenomenon
of market activities; allowance of
distributed simulations
with many com-
puters interacting through a network as well as local-host, extremely fast simulations;
possibility to design experiments mixing
human beings and artificial traders
.
2.1
Distinctive Features
Following distinctive aspects of ATOM can be highlighted:
1. ATOM can use various kind of sophisticated agents with their own behaviours and
intelligence (see section 4), including portfolio optimizers as this platform provides
the multi-assets (multi-orderbook) organisation. In addition, ATOM contains all
variety of the Euronext-Nyse orders: limit, market, stop-limit,iceberg, etc.
2. Thousands of agents can evolve simultaneously, creating a truly heterogeneous
population. Once designed, agents evolve by themselves, learning and adapting to
their (financial) environment. 100,000 agents have been employed simultaneously
for technical analysis evaluation research [7].
3. ATOM can combine human-beings and artificial traders on a single market using
its network capabilities. This feature support a wide variety of configurations: from
”experimental finance” classrooms with students, to competing strategies run inde-
pendently and distantly by several banks or research labs. The scheduler can be set
so to allow human agents to freeze the market during their decision process or not
(see above, section 3.4).
4. ATOM serves as a ”replay-engine”, meaning that it is possible to re-execute whole
trading log file with following information: identification of order topic, that corre-
sponds to stock identity, agents identification, the platform time stamp fixed at the
very moment orders arrive to a given order-book, prices resulting from the orders.
ATOM takes less than 5 seconds to replay an entire day of trading, that contains
400,000 activities. This tool is extremely important for policy-oriented experiments
focusing on the technical features of the market microstructure (tick size, price fix-
ing protocol) and its influence on the price dynamic.
