MFRS: A Multi-Frequency Reference Series Approach to Scalable and Accurate Time-Series Forecasting
Fuente:
arXiv
Saved in:
| Main Authors: | , , |
|---|---|
| Format: | Preprint |
| Published: |
2025
|
| Subjects: | |
| Online Access: | |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866929753179553792 |
|---|---|
| author | Yu, Liang Tu, Lai Bai, Xiang |
| author_facet | Yu, Liang Tu, Lai Bai, Xiang |
| contents | Multivariate time-series forecasting holds immense value across diverse applications, requiring methods to effectively capture complex temporal and inter-variable dynamics. A key challenge lies in uncovering the intrinsic patterns that govern predictability, beyond conventional designs, focusing on network architectures to explore latent relationships or temporal dependencies. Inspired by signal decomposition, this paper posits that time series predictability is derived from periodic characteristics at different frequencies. Consequently, we propose a novel time series forecasting method based on multi-frequency reference series correlation analysis. Through spectral analysis on long-term training data, we identify dominant spectral components and their harmonics to design base-pattern reference series. Unlike signal decomposition, which represents the original series as a linear combination of basis signals, our method uses a transformer model to compute cross-attention between the original series and reference series, capturing essential features for forecasting. Experiments on major open and synthetic datasets show state-of-the-art performance. Furthermore, by focusing on attention with a small number of reference series rather than pairwise variable attention, our method ensures scalability and broad applicability. The source code is available at: https://github.com/yuliang555/MFRS |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2503_08328 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | MFRS: A Multi-Frequency Reference Series Approach to Scalable and Accurate Time-Series Forecasting Yu, Liang Tu, Lai Bai, Xiang Machine Learning Information Retrieval Multivariate time-series forecasting holds immense value across diverse applications, requiring methods to effectively capture complex temporal and inter-variable dynamics. A key challenge lies in uncovering the intrinsic patterns that govern predictability, beyond conventional designs, focusing on network architectures to explore latent relationships or temporal dependencies. Inspired by signal decomposition, this paper posits that time series predictability is derived from periodic characteristics at different frequencies. Consequently, we propose a novel time series forecasting method based on multi-frequency reference series correlation analysis. Through spectral analysis on long-term training data, we identify dominant spectral components and their harmonics to design base-pattern reference series. Unlike signal decomposition, which represents the original series as a linear combination of basis signals, our method uses a transformer model to compute cross-attention between the original series and reference series, capturing essential features for forecasting. Experiments on major open and synthetic datasets show state-of-the-art performance. Furthermore, by focusing on attention with a small number of reference series rather than pairwise variable attention, our method ensures scalability and broad applicability. The source code is available at: https://github.com/yuliang555/MFRS |
| title | MFRS: A Multi-Frequency Reference Series Approach to Scalable and Accurate Time-Series Forecasting |
| topic | Machine Learning Information Retrieval |
| url | https://arxiv.org/abs/2503.08328 |