2019/07/31 by Magnus Wiese, Robert Knobloch, Ralf Korn +1 · 4 citations
Economics, Econometrics and Finance · Computer Science · Mathematics · #q-fin.MF #cs.LG #q-fin.CP #stat.ML
paper · pdf · doi:10.1080/14697688.2020.1730426
published as Quantitative Finance, 2020 · Corrected typos. Added section 2 as an overview of existing literature. Added section 5.3 to clarify the modeling assumptions. Appendix B now contains more details on the neural network architectures used. Changed latex template
arxiv created 2019/12/21 · arxiv updated 2020/04/07
Modeling financial time series by stochastic processes is a challenging task and a central area of research in financial mathematics. As an alternative, we introduce Quant GANs, a data-driven model which is inspired by the recent success of generative adversarial networks (GANs). Quant GANs consist of a generator and discriminator function, which utilize temporal convolutional networks (TCNs) and thereby achieve to capture long-range dependencies such as the presence of volatility clusters. The generator function is explicitly constructed such that the induced stochastic process allows a transition to its risk-neutral distribution. Our numerical results highlight that distributional properties for small and large lags are in an excellent agreement and dependence properties such as volatility clusters, leverage effects, and serial autocorrelations can be generated by the generator function of Quant GANs, demonstrably in high fidelity.