Validation and Calibration in ACE Models: An Investigation on the CATS model

In this paper we deal with some validation (and a …rst calibration) experiments on the CATS model proposed in Gallegati et al. (2003a, 2004b). The CATS model has been intensively used (see, for example, Delli Gatti et al., 2004; Russo, 2004; Gallegati et al., 2003b) to replicate a large number of scaling type stylized facts with a remarkable degree of precision and, for these purposes, the simulation of the model has been performed entering ad hoc parameter values and using the same initial set up for all the agents involved in the experiments. Nowadays alternative robust and reliable validation techniques for determining whether the simulation model is an acceptable representation of the real system are available (Sargent, 1996; Kleijnen, 1999); furthermore many distributional and goodness-of-…t tests have been developed (see, for example, Prabhakar, 2003; Kleiber and Kotz, 2003) while several graphical tools have been proposed to give the researcher a quick comprehension of actual and simulated data (Embrechts, 1997). This paper discusses some validation experiments performed with the CATS model. In particular starting from a sample of Italian …rms included in the AIDA database, we perform several ex-post validation experiments over the simulation period 1996-2001. In the experiments, the model parameters have been estimated using actual data and the initial set up consists of a sample of agents in 1996. The CATS model is then simulated over the period 1996-2001. Using alternative validation techniques, the simulations’ results are ex-post validated respect to the actual data. The results we achieve seem to be quite promising. JEL classi…cation: C15, C16, D31, E37, L11, L16, O31.