Fin-GAN: Forecasting and Classifying Financial Time Series via Generative Adversarial Networks
Milena Vuletić, Mihai Cucuringu and Felix Prenzel
We investigate the use of Generative Adversarial Networks (GANs) for probabilistic forecasting of financial time series. To this end, we introduce a novel economics-driven loss function for the generator. This newly designed loss function renders GANs more suitable for a classification task, and places them into a supervised learning setting, whilst producing full conditional probability distributions of price returns given previous historical values. Our approach moves beyond the point estimates traditionally employed in the forecasting literature, and allows for uncertainty estimates. Numerical experiments on equity data showcase the effectiveness of our proposed methodology, which achieves higher Sharpe Ratios compared to classical supervised learning models, such as LSTMs and ARIMA.
You are now leaving Man Group’s website
You are leaving Man Group’s website and entering a third-party website that is not controlled, maintained, or monitored by Man Group. Man Group is not responsible for the content or availability of the third-party website. By leaving Man Group’s website, you will be subject to the third-party website’s terms, policies and/or notices, including those related to privacy and security, as applicable.