FreDo: Frequency Domain-based Long-Term Time Series Forecasting
This addresses forecasting challenges in applications like climatology and energy, but it is incremental as it builds on existing baseline methods.
The paper tackles the problem of long-term time series forecasting by showing that sophisticated models may not outperform simple baselines due to error accumulation, and proposes FreDo, a frequency domain-based neural network that greatly outperforms state-of-the-art models.
The ability to forecast far into the future is highly beneficial to many applications, including but not limited to climatology, energy consumption, and logistics. However, due to noise or measurement error, it is questionable how far into the future one can reasonably predict. In this paper, we first mathematically show that due to error accumulation, sophisticated models might not outperform baseline models for long-term forecasting. To demonstrate, we show that a non-parametric baseline model based on periodicity can actually achieve comparable performance to a state-of-the-art Transformer-based model on various datasets. We further propose FreDo, a frequency domain-based neural network model that is built on top of the baseline model to enhance its performance and which greatly outperforms the state-of-the-art model. Finally, we validate that the frequency domain is indeed better by comparing univariate models trained in the frequency v.s. time domain.