Scenario generation for market risk models using generative neural networks

In this research, we show how to expand existing approaches of using generative adversarial networks (GANs) as economic scenario generators (ESG) to a whole internal model - with enough risk factors to model the full band-width of investments for an insurance company and for a one year time horizon as required in Solvency 2. For validation of this approach as well as for optimization of GAN architecture, we provide a consistent, data-driven framework using existing evaluation measures based on nearest neighbor distances and a newly developed measure for the detection of the memorizing effect. Finally, we demonstrate that the results of a GAN-based ESG are similar to regulatory approved internal models in Europe. Therefore, GAN-based models can be seen as an assumption-free data-driven alternative way of market risk modeling.