Using generative adversarial networks to synthesize artificial financial datasets
arXiv:2002.02271
Abstract
Generative Adversarial Networks (GANs) became very popular for generation of realistically looking images. In this paper, we propose to use GANs to synthesize artificial financial data for research and benchmarking purposes. We test this approach on three American Express datasets, and show that properly trained GANs can replicate these datasets with high fidelity. For our experiments, we define a novel type of GAN, and suggest methods for data preprocessing that allow good training and testing performance of GANs. We also discuss methods for evaluating the quality of generated data, and their comparison with the original real data.
References in corpus (8)
- Conditional Generative Adversarial Nets
- Evaluation: from precision, recall and F-measure to ROC, informedness, markedness and correlation
- The Effectiveness of Data Augmentation in Image Classification using Deep Learning
- Improved Training of Wasserstein GANs
- On Convergence and Stability of GANs
- Data Augmentation in Emotion Classification Using Generative Adversarial Networks
- eCommerceGAN : A Generative Adversarial Network for E-commerce
- One-Class Adversarial Nets for Fraud Detection
Cited by in corpus (6)
- Sequential Deep Learning for Credit Risk Monitoring with Tabular Financial Data
- Deep Learning Based on Generative Adversarial and Convolutional Neural Networks for Financial Time Series Predictions
- Wasserstein GAN: Deep Generation applied on Bitcoins financial time series
- Accurate and Intuitive Contextual Explanations using Linear Model Trees
- An overall view of key problems in algorithmic trading and recent progress
- Modelling and simulating spatial extremes by combining extreme value theory with generative adversarial networks