Intervention-Aware Forecasting: Breaking Historical Limits from a System Perspective

TL;DR

This paper introduces an intervention-aware forecasting framework that explicitly models external interventions, especially textual ones, to improve accuracy beyond traditional methods limited by historical data reliance.

Contribution

It proposes a novel IATSF framework and FIATS model that incorporate textual interventions using CASM and CAPS mechanisms, advancing forecasting accuracy.

Findings

FIATS outperforms state-of-the-art methods in various scenarios.

Modeling external interventions improves forecasting accuracy.

Textual interventions provide valuable qualitative information for forecasting.

Abstract

Traditional time series forecasting methods predominantly rely on historical data patterns, neglecting external interventions that significantly shape future dynamics. Through control-theoretic analysis, we show that the implicit "self-stimulation" assumption limits the accuracy of these forecasts. To overcome this limitation, we propose an Intervention-Aware Time Series Forecasting (IATSF) framework explicitly designed to incorporate external interventions. We particularly emphasize textual interventions due to their unique capability to represent qualitative or uncertain influences inadequately captured by conventional exogenous variables. We propose a leak-free benchmark composed of temporally synchronized textual intervention data across synthetic and real-world scenarios. To rigorously evaluate IATSF, we develop FIATS, a lightweight forecasting model that integrates textual interventions through Channel-Aware Adaptive Sensitivity Modeling (CASM) and Channel-Aware Parameter Sharing (CAPS) mechanisms, enabling the model to adjust its sensitivity to interventions and historical data in a channel-specific manner. Extensive empirical evaluations confirm that FIATS surpasses state-of-the-art methods, highlighting that forecasting improvements stem explicitly from modeling external interventions rather than increased model complexity alone.