CATS: Enhancing Multivariate Time Series Forecasting by Constructing Auxiliary Time Series as Exogenous Variables
PDF Proceedings arXiv Code Poster
Constructs Auxiliary Time Series (ATS) as exogenous inputs to capture inter‑series relations; identifies continuity, sparsity, and variability principles; improves multivariate TSF even with simple predictors.
A modular approach that augments original series with ATS to model cross‑series structure efficiently, yielding strong accuracy/complexity trade‑offs.