Data augmentation of credit default swap transactions based on a sequence GAN
作者：
Highlights：
• We can use the GAN model to generate more similar order data based on limited data currently available.
• We will design some evaluation criteria so that we can measure the outcome of our simulated data.
• We will show results after training to test whether our new data has the same behaviors as the original data does, which proves that our model could really help simulate limit order data.
摘要
•We can use the GAN model to generate more similar order data based on limited data currently available.•We will design some evaluation criteria so that we can measure the outcome of our simulated data.•We will show results after training to test whether our new data has the same behaviors as the original data does, which proves that our model could really help simulate limit order data.
论文关键词：Generative adversarial network,Synthetic ranking agreement index,Credit default swap transaction
论文评审过程：Received 3 August 2021, Revised 24 November 2021, Accepted 28 January 2022, Available online 14 February 2022, Version of Record 14 February 2022.
论文官网地址：https://doi.org/10.1016/j.ipm.2022.102889
