/fingan

The implementation of "modeling financial time-series with generative adversarial networks"

Primary LanguagePython

fingan

The implementation of "modeling financial time-series with generative adversarial networks" The numpy files in generated_samples folder are the time-series generated by the GAN model. To train the model, you need the data of the financial time-series. Due to the copyright issue, we can not publish it here. Please contact shushakura(at)gmail.com for the information.

bibliography

Shuntaro Takahashi, Yu Chen and Kumiko Tanaka-Ishii. Modeling financial time-series with generative adversarial networks. Physica A. 527, 121261, 2019

requirement

  • Python 3.6
  • tensorflow-gpu 1.14
  • matplotlib 3.1.1
  • pandas 0.25.1
  • numpy 1.17.1