Position: Beyond Model-Centric Prediction -- Agentic Time Series Forecasting
This is a foundational position paper that could influence future research in time series forecasting by shifting from static models to adaptive agentic systems.
The paper argues that traditional model-centric time series forecasting is insufficient for adaptive settings and proposes agentic time series forecasting (ATSF), which reframes forecasting as an agentic process involving perception, planning, action, reflection, and memory to enable interactive and iterative workflows.
Time series forecasting has traditionally been formulated as a model-centric, static, and single-pass prediction problem that maps historical observations to future values. While this paradigm has driven substantial progress, it proves insufficient in adaptive and multi-turn settings where forecasting requires informative feature extraction, reasoning-driven inference, iterative refinement, and continual adaptation over time. In this paper, we argue for agentic time series forecasting (ATSF), which reframes forecasting as an agentic process composed of perception, planning, action, reflection, and memory. Rather than focusing solely on predictive models, ATSF emphasizes organizing forecasting as an agentic workflow that can interact with tools, incorporate feedback from outcomes, and evolve through experience accumulation. We outline three representative implementation paradigms -- workflow-based design, agentic reinforcement learning, and a hybrid agentic workflow paradigm -- and discuss the opportunities and challenges that arise when shifting from model-centric prediction to agentic forecasting. Together, this position aims to establish agentic forecasting as a foundation for future research at the intersection of time series forecasting.