Creating Disasters: Recession Forecasting with GAN-Generated Synthetic Time Series Data

TL;DR

This paper demonstrates that using GAN-generated synthetic time series data can enhance the accuracy of recession forecasting and Treasury yield predictions, addressing data scarcity issues in financial modeling.

Contribution

It introduces DoppelGANger, a GAN model for generating synthetic time series data, and shows its effectiveness in improving recession and yield forecasts.

Findings

Synthetic data improves short-term Treasury yield forecasting.

Training on synthetic recessions enhances future recession prediction.

Synthetic data can supplement limited real data for better financial models.

Abstract

A common problem when forecasting rare events, such as recessions, is limited data availability. Recent advancements in deep learning and generative adversarial networks (GANs) make it possible to produce high-fidelity synthetic data in large quantities. This paper uses a model called DoppelGANger, a GAN tailored to producing synthetic time series data, to generate synthetic Treasury yield time series and associated recession indicators. It is then shown that short-range forecasting performance for Treasury yields is improved for models trained on synthetic data relative to models trained only on real data. Finally, synthetic recession conditions are produced and used to train classification models to predict the probability of a future recession. It is shown that training models on synthetic recessions can improve a model's ability to predict future recessions over a model trained only on real data.