Fiaingen: A financial time series generative method matching real-world data quality
This addresses data limitations for financial machine learning applications, such as trading and investment, but is incremental as it builds on existing generative methods.
The paper tackles the shortage of high-quality financial time series data by introducing Fiaingen, a generative method that produces synthetic data closely matching real-world data in reduced dimensionality space, downstream task performance, and runtime, with generation times near seconds and models achieving performance close to those trained on real data.
Data is vital in enabling machine learning models to advance research and practical applications in finance, where accurate and robust models are essential for investment and trading decision-making. However, real-world data is limited despite its quantity, quality, and variety. The data shortage of various financial assets directly hinders the performance of machine learning models designed to trade and invest in these assets. Generative methods can mitigate this shortage. In this paper, we introduce a set of novel techniques for time series data generation (we name them Fiaingen) and assess their performance across three criteria: (a) overlap of real-world and synthetic data on a reduced dimensionality space, (b) performance on downstream machine learning tasks, and (c) runtime performance. Our experiments demonstrate that the methods achieve state-of-the-art performance across the three criteria listed above. Synthetic data generated with Fiaingen methods more closely mirrors the original time series data while keeping data generation time close to seconds - ensuring the scalability of the proposed approach. Furthermore, models trained on it achieve performance close to those trained with real-world data.