In the simulations time has to be advanced as well as events have to be handled, so they have to be scheduled in order to let them happen.

Simulation literature shows several different scheduler mechanism focused on different authors' needs hardly well documented or related to computer science literature. The correctness of the scheduling process and the reliability of the experiment repeatability are critical point in agent based models.

In this work we discuss several classical scheduler schemes, focusing on the experimental comparison about the usage of two methods driven by opposite philosophical approaches: the “central static” (inspired by Langton, Minar and Burkhart, 1996) and “dynamic” (inspired to Franta Maly, 1977) schedulers.

The first one is top-down oriented, so the scheduler chooses who's doing what and when too; Each agent will be activated according to a sequential list.

In the second a bottom up strategy is applied: each agent can freely decide when they will be activated.

Their application, in most financial market models, shows the same statistical properties considering returns distributions or price and volumes dynamics. However some specific research question can be better answered by one of the two methods in terms of efforts saving in implementation, higher code comprehension and runtime performances.

In fact we found that to study a treasury auctions (primary debt market), or a blocks market, in which every agent makes a proposal without any direct interaction among the others, a “central static” meets our need.

Conversely to describe agent reactions to news or to price falling down “speed” in a continuous double auction market, we preferred a dynamic scheduling, in which any agent can freely react to information flows.

The cited experiments are done using the SUM/SumWEB platform (Terna 2000, Cappellini and Ferraris 2008), able to face different market institutions as well as different scheduling mechanisms.

• Legal notice:
  

  Copyright (c) Pielaszek Research, all rights reserved.
  The above materials, including auxiliary resources, are subject to Publisher's copyright and the Author(s) intellectual rights. Without limiting Author(s) rights under respective Copyright Transfer Agreement, no part of the above documents may be reproduced without the express written permission of Pielaszek Research, the Publisher. Express permission from the Author(s) is required to use the above materials for academic purposes, such as lectures or scientific presentations.
  In every case, proper references including Author(s) name(s) and URL of this webpage: https://science24.com/paper/14866 must be provided.