Abstract:The non-stationary nature of real-world Multivariate Time Series (MTS) data presents forecasting models with a formidable challenge of the time-variant distribution of time series, referred to as distribution shift. Existing studies on the distribution shift mostly adhere to adaptive normalization techniques for alleviating temporal mean and covariance shifts or time-variant modeling for capturing temporal shifts. Despite improving model generalization, these normalization-based methods often assume a time-invariant transition between outputs and inputs but disregard specific intra-/inter-series correlations, while time-variant models overlook the intrinsic causes of the distribution shift. This limits model expressiveness and interpretability of tackling the distribution shift for MTS forecasting. To mitigate such a dilemma, we present a unified Probabilistic Graphical Model to Jointly capturing intra-/inter-series correlations and modeling the time-variant transitional distribution, and instantiate a neural framework called JointPGM for non-stationary MTS forecasting. Specifically, JointPGM first employs multiple Fourier basis functions to learn dynamic time factors and designs two distinct learners: intra-series and inter-series learners. The intra-series learner effectively captures temporal dynamics by utilizing temporal gates, while the inter-series learner explicitly models spatial dynamics through multi-hop propagation, incorporating Gumbel-softmax sampling. These two types of series dynamics are subsequently fused into a latent variable, which is inversely employed to infer time factors, generate final prediction, and perform reconstruction. We validate the effectiveness and efficiency of JointPGM through extensive experiments on six highly non-stationary MTS datasets, achieving state-of-the-art forecasting performance of MTS forecasting.

A robust functional time series forecasting method

On the modelling and prediction of high-dimensional functional time series

Graphical Principal Component Analysis of Multivariate Functional Time Series

Functional volatility forecasting

Dynamic principal component regression for forecasting functional time series in a group structure

High-dimensional functional time series forecasting: An application to age-specific mortality rates

DYNAMIC PRINCIPAL COMPONENT REGRESSION: APPLICATION TO AGE-SPECIFIC MORTALITY FORECASTING

Robust Multi-Dimensional Time Series Forecasting

Ensemble Modeling for Time Series Forecasting: an Adaptive Robust Optimization Approach

Functional time series forecasting of extreme values

Adaptive Robust Polynomial Regression for Power Curve Modeling with Application to Wind Power Forecasting

Robust Function-on-Function Regression

Robust functional regression based on principal components

Fully Automated Correlated Time Series Forecasting in Minutes

Bayesian Forecasting of Bounded Poisson Distributed Time Series

Functional principal component analysis for longitudinal observations with sampling at random

Robust Multivariate Time Series Forecasting against Intra- and Inter-Series Transitional Shift

Bayesian nonparametric forecasting of monotonic functional time series

Dynamic functional time-series forecasts of foreign exchange implied volatility surfaces

Shape-Preserving Prediction for Stationary Functional Time Series

Robust Bayesian Functional Principal Component Analysis