Genhack_competition

Introduction

Stress tests have become a main guideline for the regulator in order to assess the banking system resilience against the realizations of various categories of risk (market, credit, operational, climate, etc). The main challenge is to simulate unfavorable extreme (but plausible) negative returns similar to a historical dataset.

Submission

In general,

• Function G (the model)
• The associated trained parameters theta
• The sampled Z (csv file with 410 or 408 rows and your chosen dimension of Z)
  • Also, it must be sampled from a standard Gaussian
• The outputs (csv file with 410 or 408 rows and 4 columns)

The zipped folder should include,

Along with the simulated data in a .csv file of syntax <guild-name>number_submission.csv, you will also submit:

• Simulated data generated_samples.csv
• The noise values with which you simulated the data noise.csv
• The python code of the inference
  • main.py file
  • requirements.txt (all dependencies)
• paramaters of the model

Resources

• Different methods to generate data

Useful links for GAN

• To create GAN using PyTorch
• To create GAN using TensorFlow
• Basics of GAN bu google
• Neural networks architecture
• Generative models in Finance
• From GAN to WGAN