TimeCatcher: A Variational Framework for Volatility-Aware Forecasting of Non-Stationary Time Series

TL;DR

TimeCatcher introduces a volatility-aware variational framework that enhances long-term forecasting of non-stationary time series by capturing hidden dynamics and significant local variations, outperforming existing models across diverse real-world datasets.

Contribution

The paper presents a novel framework combining variational encoding and volatility awareness to improve long-term forecasting of highly non-stationary time series.

Findings

Outperforms state-of-the-art models on nine real-world datasets.

Significant improvements in long-term forecasts with high volatility.

Effective in domains like traffic, financial, energy, and weather data.

Abstract

Recent lightweight MLP-based models have achieved strong performance in time series forecasting by capturing stable trends and seasonal patterns. However, their effectiveness hinges on an implicit assumption of local stationarity assumption, making them prone to errors in long-term forecasting of highly non-stationary series, especially when abrupt fluctuations occur, a common challenge in domains like web traffic monitoring. To overcome this limitation, we propose TimeCatcher, a novel Volatility-Aware Variational Forecasting framework. TimeCatcher extends linear architectures with a variational encoder to capture latent dynamic patterns hidden in historical data and a volatility-aware enhancement mechanism to detect and amplify significant local variations. Experiments on nine real-world datasets from traffic, financial, energy, and weather domains show that TimeCatcher consistently outperforms state-of-the-art baselines, with particularly large improvements in long-term forecasting scenarios characterized by high volatility and sudden fluctuations. Our code is available at https://github.com/ColaPrinceCHEN/TimeCatcher.