The econometric model used in LUCAS is a dynamic, stochastic
model which, by definition, has at least one random variable, namely
land cover, and deals explicitly with time-variable interaction. A
stochastic simulation
uses a statistical
sampling of multiple replicates, repeated simulations, of the same
model. Simulations are used when the systems being modeled are too
complex to be described by a set of mathematical equations for which
an analytic solution is attainable. Simulation is, however, an
imprecise technique and provides only statistical estimates and not
exact results. The LUCAS computer simulation serves as a tool to help
evaluate land-use management policies before actually implementing
them on the real system.
An effective model need only have a high correlation between its prediction and the actual outcome in the real system, not necessarily approximate the real system, so statistical results are sufficient to understand the results of the simulated model [29]. Therefore, the primary output produced by LUCAS are statistics collected during the simulation which can be viewed as a graph or analyzed with a statistical package, such as SAS [30]. Figure 2 outlines the modular model implemented in LUCAS; the validation of which is discussed in Section 5.1. Aside from predicting future behavior, models also allow investigators to explore the nature of potential scenarios. Each module of the LUCAS model is described in detail in the following sections.