Sign In
insight - Forecasting
Rethinking Channel Dependence for Multivariate Time Series Forecasting: Learning from Leading Indicators
Exploiting locally stationary lead-lag relationships between variates improves multivariate time series forecasting accuracy.
Hierarchical Time Series Variational Transformer for Nonstationary Multivariate Time Series Forecasting
HTV-Trans is a novel model that effectively captures non-stationarity and stochasticity in multivariate time series forecasting, outperforming other methods.
TimeXer: Empowering Transformers for Time Series Forecasting with Exogenous Variables
TimeXer proposes a novel framework to enhance time series forecasting by reconciling endogenous and exogenous information using the Transformer architecture. It achieves state-of-the-art performance in various real-world datasets.
Products
© 2024 by Linnk AI