Marco Valente

''Keep calm and carry on''

  • Increase font size
  • Default font size
  • Decrease font size
Home Simulations SIME Course


Simulation in Evolutionary Economics

SIME training course 

Supported by the DIME Network of Excellence of the European Union

Strasbourg, France, May 10th to May 15th, 2009


Deadline for application: March 15th, 2009



The course aims at teaching students how to use simulation models for research purposes, particularly agent-based ones used in evolutionary models. A simulation project entails different steps: design, coding, initialization, data analysis, presentation of results and extensions. The intrinsic uncertainty in developing a model for research implies that a simulation project cannot develop linearly, but frequently requires revising prior work, even radically, making simulation development more difficult than a standard programming project. Moreover, the lack of an accepted methodology makes unclear how one should assess scientific results obtained by simulations. Therefore, the skills and protocols required in developing simulation models for research purposes differ from those required for general software development.

The course addresses all the issues involved in the use of simulation for scientific purposes, from the basic programming knowledge required to express common dynamics up to the epistemological relevance of simulation results.


The course is based on the development tool Laboratory for Simulation Development – LSD. Event though LSD is essentially a programming language, it allows users to concentrate their time on the scientific content of the simulation model, while the technical aspects required by the implementation are automatically managed by the system. In practice LSD users express their models by defining a set of variables as in a system of difference equations, and need only to write the code expressing the computation for the variables. These chunks of code are automatically assembled by the system to actually perform a simulation step, with the user controlling the model via graphical interfaces in order, e.g. to get full access to the model data, represent graphically the results, being signaled errors, etc. The intrinsic modularity of LSD, coupled with the clear separation between code, data and system interfaces favor the application of a gradual approach to development, starting from simple prototyping and proceeding by adding new elements to the model.

LSD is particularly suited for large agent-based models, where the automatic management of issues such as the priority of computation, large number of elements and speed are crucial. LSD is implemented in C++ and is available for MS Windows, Mac OS, and Linux.


Doctoral students and researchers working on or in need of a simulation model. No previous experience  on simulation or programming is required, though the course is of interest also to expert modellers willing to compare the LSD approach with other simulation languages. Applicants with specific interests (e.g. interested in specific models) will be given individually tailored tutoring.


The course (5 full-time days) consists of lectures on the use of LSD, the methodology of simulations and extensive exercises during which students learn how to design, implement, revise and generally exploit simulation models for scientific research. Several models, of increasing complexity, will be developed as exercise. Students will be invited to develop their own original modification of one of the model presented, or to design a new model. Member of the staff will assist students during the exercises and provide support tailored on the prior skills of individual students. Participants are expected to bring their own computers.


P. LLERENA, BETA, Strasbourg, F; M. VALENTE, University of L’Aquila & LEM, Pisa, I.


T. CIARLI, Max Planck Institute for Economics, Jena, D; A. LORENTZ, Max Planck Institute for Economics, Jena, D; M. VALENTE, University of L’Aquila & LEM Pisa, I.


Fees : 400 euros for the May session (200 euros for participants from DIME partner institutions). Accommodation and other local costs will be covered by the DIME budget.

Travel: participants are requested to look for local funds.  If necessary they can apply for support from DIME; limited funds will be available and will be used primarily for applicants from isolated teams and institutions from new member states as well as for women.

Application procedure:

Applications should be sent (by e-mail  if possible) to Patrick Llerena *

- Letter of motivation with CV

- Abstract of PhD topic, or current research area

- Letter of support by the home supervisor

- In case the applicant is interested in a specific model, the application should also include a

description of the project, including the aim of the simulation and current stage of development.



* Patrick Llerena – Université de Strasbourg – BETA- 61 avenue de la Forêt Noire – 67000 Strasbourg (France) – Fax : 33(0)3 90242071 - E-mail : This e-mail address is being protected from spambots. You need JavaScript enabled to view it  


Last Updated on Wednesday, 25 February 2009 13:53