Frequency Adaptive Normalization For Non-stationary Time Series Forecasting
This work provides an incremental improvement in time series forecasting accuracy for practitioners dealing with non-stationary data.
This paper tackles the problem of non-stationary time series forecasting, which involves evolving trend and seasonal patterns. The authors propose Frequency Adaptive Normalization (FAN), a model-agnostic method that extends instance normalization to handle both dynamic trend and seasonal patterns by employing Fourier transform to identify predominant frequency components. FAN demonstrates significant performance advancement, achieving 7.76% to 37.90% average improvements in MSE across eight benchmark datasets.
Time series forecasting typically needs to address non-stationary data with evolving trend and seasonal patterns. To address the non-stationarity, reversible instance normalization has been recently proposed to alleviate impacts from the trend with certain statistical measures, e.g., mean and variance. Although they demonstrate improved predictive accuracy, they are limited to expressing basic trends and are incapable of handling seasonal patterns. To address this limitation, this paper proposes a new instance normalization solution, called frequency adaptive normalization (FAN), which extends instance normalization in handling both dynamic trend and seasonal patterns. Specifically, we employ the Fourier transform to identify instance-wise predominant frequent components that cover most non-stationary factors. Furthermore, the discrepancy of those frequency components between inputs and outputs is explicitly modeled as a prediction task with a simple MLP model. FAN is a model-agnostic method that can be applied to arbitrary predictive backbones. We instantiate FAN on four widely used forecasting models as the backbone and evaluate their prediction performance improvements on eight benchmark datasets. FAN demonstrates significant performance advancement, achieving 7.76% ~ 37.90% average improvements in MSE.