FracTS generates diverse long-range multivariate time series efficiently
FracTS: Hierarchical and Autoregressive Time Series Generation
Machine Learning
Summary
Time series data, which tracks how things change over time, is important in many fields but can be hard to produce realistically. The authors present FracTS, a new way to create synthetic time series that captures both overall trends and fine details by working from coarse to fine scales. It improves over older methods by better handling complex time relationships and creating more varied, accurate sequences. The authors tested FracTS on multiple datasets and found it consistently made higher quality synthetic data than other approaches.
What this means in practice
- •For financial data analysts: Create realistic synthetic financial time series for risk testing without exposing sensitive real data.
- •For healthcare data engineers: Generate multivariate patient monitor data with improved temporal coherence for algorithm training while preserving privacy.
Authors
Jiayu Li, Umair Afzal, Zilong Zhao, Milad Abdollahzadeh, Uzair Javaid, Biplab Sikdar
Abstract
Time series are ubiquitous across different domains, and the growing use of synthetic data for privacy-preserving sharing, data augmentation, and stress testing makes time series generation increasingly important. Early generative approaches are predominantly autoregressive, which can suffer from low sample diversity, error accumulation, and difficulty modeling long-range dependencies. In this paper, we introduce FracTS, a hierarchical, coarse-to-fine autoregressive model for multivariate time series generation. Inspired by fractal generative modeling developed for images, FracTS adapts the paradigm from 2D to 1D sequences by learning a multi-scale representation that captures global trajectory structure at coarse scales and local dynamics at fine scales while following the autoregressive nature of time series. In contrast to unstructured images, time series are structured, allowing easy automated or designated temporal feature processing optimized for the model, and global (static) feature extraction describing the instances that generate the corresponding time series in the collection. We incorporate them in the generation process to further improve the generation quality. The design naturally handles multivariate dependencies and improves long-horizon temporal coherence without requiring fixed-length inputs. Experiments on five datasets (two conditional, three unconditional) conducted against seven baselines show that FracTS consistently outperforms baselines across marginal distribution and predictive ML utility, and correlation preservation in the unconditional setting. In particular, the predictive ML utility is averagely improved by 100.0% and 75.0-81.0% for unconditional and conditional time series datasets compared to the best baseline (relative error computed on dataset-averaged raw values). Code is available at: https://github.com/BetterdataLabs/fracts.