NatLab, an asynchronous simulation platform for artificial stock markets

NatLab www.complexity-research.org/natlab

NatLab is an asynchronous simulation platform for artificial stock markets.
This project was initiated by Sorin Solomon and Lev Muchnik,
and is currently maintained by Gilles Daniel.

Every collaboration is welcome, so please do not hesitate to contact us.

Download

The source code for this project is not available yet -- we are working on it. Nevertheless, the application itself can be downloaded and simulations can be run.

Development Status: 1.5 (June 2006)
Intended Audience: Financial economists, Physicists, Computer Scientists
Operating System: Windows 95/98/NT/2000/XP

NatLab platform

	- NatLab-1.5	v1.5, June 2006, 880KB
	- NatLab-1.4	v1.4, April 2006, 860KB
	- NatLab-1.3	v1.3, March 2006, 1450KB
	- NatLab-1.2	v1.2, March 2006, 890KB
	- NatLab-1.1	v1.1, January 2006, 910KB
	- NatLab-1.0	v1.0, June 2004, 911KB

Agents

	- randomAgents	Zero-intelligence agents
	- allAgents-20060129	All agents as on January 29th, 2006

Example of output

- /output

Price dynamics, order book, etc.

Matlab post-processing

	- FiStyL library	Matlab scripts to analyse simulated data
	- JobScheduler-v2	Generate some configuration files to run batch jobs

Documentation

	- Design principles	Functionalities of the NatLab platform
	- Markov Webs	Theory underlying the asynchronous engine

Here is the changelog.

These are a few screenshots of NatLab simulation platform.
Click on the picture to enlarge.

Design the market	Select the traders
Run a simulation

Publications

The following papers have been published about NatLab itself or results obtained from simulations generated with this platform.

Lev Muchnik and Sorin Solomon (2006)
Markov Nets and the NatLab platform; Application to Continuous Double Auction
To appear in Proc. Vol. to be published by Springer in its series "New Economic Windows"

[ PDF ]

Gilles Daniel, Lev Muchnik and Sorin Solomon (2005)
Traders imprint themselves by adaptively updating their own avatar
In proceedings of Artificial Economics 2005, Lille, France (2005), 27-38

[ PDF ]

Lev Muchnik, Yoram Louzoun and Sorin Solomon (2004)
Agent Based Simulation Design Principles - Applications to Stock Market
To appear in: Practical Fruits of Econophysics, proceedings of the "Nikkei Econophysics III" conference held in Tokyo November 2004. Publisher: Springer

[ PDF ]

L. Muchnik, F. Slanina, and S. Solomon (2003)
The Interacting Gaps Model: Reconciling Theoretical and Numerical Approaches to Limit-Order Models
Physica A 330 (2003) 232.

[ PDF ]

Lev Muchnik and Sorin Solomon (2003)
Statistical Mechanics of Conventional Traders May Lead to Non-Conventional Market Behavior
Physica Scripta, Vol. T106, 41, 2003

[ PDF ]

L. Muchnik (2003)
Simulating Emergence and Complex Collective Dynamics in the Stock Markets
Masters Thesis, The Hebrew University of Jerusalem, Israel, 2003

[ PDF ]

M. Shatner, L. Muchnik, M. Leshno, and S. Solomon (2000)
A continuous time asynchronous model of the stock market; beyond the LLS model
In Economic Dynamics from the Physics Point of View, Physikzentrum Bad Honnef, Germany, 2000

[ PDF ]

S. Solomon (1999)
Behaviorly realistic simulations of stock market traders with a soul
Physics Communications, vol. 121-122 (1999) p. 161-167

[ PDF ]

Other market simulators in town

B. Jacobs, K. Levy and H. Markowitz' JLM Simulator
M. Raberto, S. Cincotti et al' Genoa Artificial Stock Market (GASM)
K. Boer, U. Kaymak et al' ABSTRACTE trading environment
SantaFe Institure' Artificial Stock Market (ASM)
For laboratory experiments: Caltech' jMarkets

Contact us

For any enquiry, please contact

Lev Muchnik levmuchnik [at] gmail.com
Gilles Daniel gilles [at] isi.it