Multi-Order Wavelet Derivative Transform for Deep Time Series Forecasting

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Zhou, Ziyu, Hu, Jiaxi, Wen, Qingsong, Kwok, James T., Liang, Yuxuan
Format: Preprint
Published: 2025
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866914309638979584
author Zhou, Ziyu
Hu, Jiaxi
Wen, Qingsong
Kwok, James T.
Liang, Yuxuan
author_facet Zhou, Ziyu
Hu, Jiaxi
Wen, Qingsong
Kwok, James T.
Liang, Yuxuan
contents In deep time series forecasting, the Fourier Transform (FT) is extensively employed for frequency representation learning. However, it often struggles in capturing multi-scale, time-sensitive patterns. Although the Wavelet Transform (WT) can capture these patterns through frequency decomposition, its coefficients are insensitive to change points in time series, leading to suboptimal modeling. To mitigate these limitations, we introduce the multi-order Wavelet Derivative Transform (WDT) grounded in the WT, enabling the extraction of time-aware patterns spanning both the overall trend and subtle fluctuations. Compared with the standard FT and WT, which model the raw series, the WDT operates on the derivative of the series, selectively magnifying rate-of-change cues and exposing abrupt regime shifts that are particularly informative for time series modeling. Practically, we embed the WDT into a multi-branch framework named WaveTS, which decomposes the input series into multi-scale time-frequency coefficients, refines them via linear layers, and reconstructs them into the time domain via the inverse WDT. Extensive experiments on ten benchmark datasets demonstrate that WaveTS achieves state-of-the-art forecasting accuracy while retaining high computational efficiency.
format Preprint
id arxiv_https___arxiv_org_abs_2505_11781
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Multi-Order Wavelet Derivative Transform for Deep Time Series Forecasting
Zhou, Ziyu
Hu, Jiaxi
Wen, Qingsong
Kwok, James T.
Liang, Yuxuan
Machine Learning
In deep time series forecasting, the Fourier Transform (FT) is extensively employed for frequency representation learning. However, it often struggles in capturing multi-scale, time-sensitive patterns. Although the Wavelet Transform (WT) can capture these patterns through frequency decomposition, its coefficients are insensitive to change points in time series, leading to suboptimal modeling. To mitigate these limitations, we introduce the multi-order Wavelet Derivative Transform (WDT) grounded in the WT, enabling the extraction of time-aware patterns spanning both the overall trend and subtle fluctuations. Compared with the standard FT and WT, which model the raw series, the WDT operates on the derivative of the series, selectively magnifying rate-of-change cues and exposing abrupt regime shifts that are particularly informative for time series modeling. Practically, we embed the WDT into a multi-branch framework named WaveTS, which decomposes the input series into multi-scale time-frequency coefficients, refines them via linear layers, and reconstructs them into the time domain via the inverse WDT. Extensive experiments on ten benchmark datasets demonstrate that WaveTS achieves state-of-the-art forecasting accuracy while retaining high computational efficiency.
title Multi-Order Wavelet Derivative Transform for Deep Time Series Forecasting
topic Machine Learning
url https://arxiv.org/abs/2505.11781