Xihao Piao, Zheng Chen, Lingwei Zhu, Yushun Dong, Yasuko Matsubara, Yasushi Sakurai
TIFO is a new method that improves time series forecasting by addressing distribution shifts using a frequency-based approach, achieving significant accuracy and efficiency gains.
Predicting future values in time series data, like stock prices or weather patterns, is challenging when the data's behavior changes over time. Traditional methods struggle because they don't fully account for these changes. This paper introduces TIFO, a new technique that focuses on the frequency components of the data to better handle these shifts. By identifying stable patterns and ignoring unstable ones, TIFO improves prediction accuracy and reduces computational effort, making it a powerful tool for forecasting in various applications.
