Does Synthetic Data Help? Empirical Evidence from Deep Learning Time Series Forecasters

TL;DR

This large-scale empirical study investigates the impact of synthetic data on deep learning time series forecasting, revealing architecture-dependent effects and practical guidelines for effective augmentation.

Contribution

It provides the first comprehensive evaluation of synthetic data augmentation in time series forecasting, highlighting when and how it improves model performance.

Findings

Channel-mixing models benefit from synthetic augmentation in most cases.

Synthetic augmentation can significantly improve low-resource forecasting performance.

Only the Seasonal-Trend generator consistently enhances results across benchmarks.

Abstract

Synthetic data has transformed language model training, yet its role in time series forecasting remains poorly understood. We present a large-scale empirical study: nine experiment groups, 4,218 runs systematically evaluating synthetic time series augmentation across five architectures, four synthetic signals and seven datasets. The effect is sharply architecture-conditional: channel-mixing models (TimesNet, iTransformer) benefit in the majority of trials, while channel-independent models (DLinear, PatchTST) are consistently degraded. In selected low-resource settings the gains are striking: TimesNet trained on only 10\% of Weather data with synthetic augmentation surpasses the full-data baseline (4 of 16 sparsity-dataset combinations). Averaged across all architectures, augmentation hurts in 67\% of trials. We further find that only the Seasonal-Trend generator reliably helps across the tested benchmarks, and that hard curriculum switching is actively harmful (+24\% MSE degradation). These results provide concrete, actionable guidelines on how to use synthetic data: use synthetic augmentation with channel-mixing architectures, use gradual annealing schedules, and treat low-resource augmentation as architecture- and dataset-dependent. Code is available at \href{https://github.com/hugoiscracked/synthetic-ts/tree/main}