Second-order Control of Complex Systems with Correlated Synthetic Data

The generation of synthetic data is an essential tool to study complex systems, allowing for example to test models of these in precisely controlled settings, or to parametrize simulation models when data is missing. This paper focuses on the generation of synthetic data with an emphasis on correlation structure. We introduce a new methodology to generate such correlated synthetic data. It is implemented in the field of socio-spatial systems, more precisely by coupling an urban growth model with a transportation network generation model. We also show the genericity of the method with an application on financial time-series. The simulation results show that the generation of correlated synthetic data for such systems is indeed feasible within a broad range of correlations, and suggest applications of such synthetic datasets.