Self-Adaptive Scale Handling for Forecasting Time Series with Scale Heterogeneity

2026-06-18

AI Summary

This work addresses the challenge posed by scale heterogeneity in real-world industrial time series, where conventional normalization methods either compress low-magnitude signals or amplify inverse scaling errors. To overcome this limitation, the authors propose an Adaptive Scaling (AS) module that dynamically determines whether to apply scale adjustment to input sequences based on their characteristics. The AS module integrates prior statistical knowledge with a neural network–driven calibration mechanism to enable input-dependent adaptive scaling, effectively avoiding over-calibration while preserving the semantic distinguishability of signals across different magnitudes. Designed for seamless integration into mainstream time series forecasting (TSF) architectures, the AS module consistently enhances prediction performance across multiple TSF models when evaluated on real-world fund sales datasets from Ant Fortune and Alipay.

Abstract

Current time series forecasting (TSF) research predominantly focuses on scale-homogeneous data, where different time series share similar numerical magnitude ranges. However, in real-world industrial scenarios such as financial product sales, different time series often differ by orders of magnitude (scale heterogeneity). Since these series share similar temporal patterns, joint modeling is desirable for better data utilization, yet existing scaling methods either compress low-scale signals (global normalization) or destroy semantic discriminability and amplify inverse-scaling errors (window-based scaling). This paper proposes a self-Adaptive Scale-handling (AS) module that learns adaptive scale factors tailored to each input, preserving semantic discriminability while reducing inverse-scaling errors. AS consists of Scale Calibrating (SC), which calibrates prior mean scaling factors through neural networks, and Scaling Selection (SS), which decides whether to apply calibration or retain the original factor, avoiding over-calibration. Experiments on real-world fund sales datasets from Ant Fortune and Alipay show that AS seamlessly integrates into popular TSF models and consistently improves their performance. The code and dataset are available at the link https://github.com/Meteor-Stars/ASTSF.

Problem

• scale heterogeneity
• time series forecasting
• scaling methods
• semantic discriminability
• inverse-scaling errors

Innovation

• scale heterogeneity
• self-adaptive scaling
• time series forecasting
• semantic discriminability
• inverse-scaling error