Network time series forecasting using spectral graph wavelet transform
Published:
Abstract
We proposed a novel method for forecasting network time series that occur in graphs or networks. Our approach is based on a spectral graph wavelet transform (SGWT) that provides the localized behavior of graph signals around each node. The proposed method improves forecasting performance over other existing methods. In particular, the advantages of the proposed method stand out when signals observed on a graph are inhomogeneous or non-stationary.
Contribution
We proposed a new method for forecasting network time series using SGWT. https://doi.org/10.1016/j.ijforecast.2023.08.006