# Benchmarking Time Series Generation Methods for Privacy-Preserving Forecasting

arXiv: https://arxiv.org/abs/2608.10891 · HTML: https://mlsanity.com/paper/2608.10891/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 12 of its 28 references that are in the map.
- First posted on arXiv 11 Aug 2026.

## Builds on (most cited first, 12 of 12)

- [Are Transformers Effective for Time Series Forecasting?](https://mlsanity.com/paper/2205.13504/) (2022)
- [Chronos: Learning the Language of Time Series](https://mlsanity.com/paper/2403.07815/) (2024)
- [TimeVAE: A Variational Auto-Encoder for Multivariate Time Series Generation](https://mlsanity.com/paper/2111.08095/) (2021)
- [Data Augmentation of Wearable Sensor Data for Parkinson's Disease Monitoring using Convolutional Neural Networks](https://mlsanity.com/paper/1706.00527/) (2017)
- [Predict, Refine, Synthesize: Self-Guiding Diffusion Models for Probabilistic Time Series Forecasting](https://mlsanity.com/paper/2307.11494/) (2023)
- [Benchmarking Differentially Private Synthetic Data Generation Algorithms](https://mlsanity.com/paper/2112.09238/) (2021)
- [Criteria for Classifying Forecasting Methods](https://mlsanity.com/paper/2212.03523/) (2022)
- [Improving the Accuracy of Global Forecasting Models using Time Series Data Augmentation](https://mlsanity.com/paper/2008.02663/) (2020)
- [Quantifying Differential Privacy under Temporal Correlations](https://mlsanity.com/paper/1610.07543/) (2016)
- [Online Data Augmentation for Forecasting with Deep Learning](https://mlsanity.com/paper/2404.16918/) (2024)
- [ModelRadar: Aspect-based Forecast Evaluation](https://mlsanity.com/paper/2504.00059/) (2025)
- [Grasynda: Graph-based Synthetic Time Series Generation](https://mlsanity.com/paper/2601.19668/) (2026)

## Abstract (arXiv)

Time series forecasting in privacy-sensitive domains often requires training models on released data rather than original observations. Synthetic time series generation has been developed primarily for data augmentation, where generated series supplement the original training set. How well these methods perform when fully replacing the original data - and how much privacy risk the released series carry - remains underexplored. We address this gap through a benchmark evaluating synthetic generation methods and noise-based anonymization baselines under a Train on Synthetic, Test on Real (TSTR) protocol. We jointly assess forecasting performance and distance-based empirical privacy risk across seven datasets, characterizing the trade-off between these objectives. We also introduce Grasynda-P, a privacy-motivated extension of the graph-based generator Grasynda, incorporating matrix ensembling and kernel density estimation. Our results show that: (1) no generation method fully substitutes for original training data; (2) noise-based anonymization yields the strongest privacy but the worst forecasting performance; (3) simple transformation-based generators outperform deep generative models for forecasting in this setting; and (4) Grasynda-P lies on the Pareto frontier, achieving competitive forecasting with stronger privacy separation than other generators. This benchmark establishes a reference point for evaluating and developing new privacy-aware synthetic time series generation methods.