AlphaCast: A Human Wisdom-LLM Intelligence Co-Reasoning Framework for Interactive Time Series Forecasting

TL;DR

AlphaCast introduces an interactive, multi-stage reasoning framework utilizing large language models for time series forecasting, mimicking human expert iterative reasoning to improve accuracy.

Contribution

It reformulates forecasting as an autonomous, multi-turn reasoning process supported by external tools, enhancing LLM capabilities without additional training.

Findings

Outperforms baseline methods on multiple benchmarks.

Supports diverse expert perspectives through external toolkits.

Enables iterative, autonomous forecasting with LLMs.

Abstract

Time series forecasting plays a crucial role in decision-making across many real-world applications. Despite substantial progress, most existing methods still treat forecasting as a static, single-pass regression problem. In contrast, human experts form predictions through iterative reasoning that integrates temporal features, domain knowledge, case-based references, and supplementary context, with continuous refinement. In this work, we propose Alphacast, an interaction-driven agentic reasoning framework that enables accurate time series forecasting with training-free large language models. Alphacast reformulates forecasting as an expert-like process and organizes it into a multi-stage workflow involving context preparation, reasoning-based generation, and reflective evaluation, transforming forecasting from a single-pass output into a multi-turn, autonomous interaction process. To support diverse perspectives commonly considered by human experts, we develop a lightweight toolkit comprising a feature set, a knowledge base, a case library, and a contextual pool that provides external support for LLM-based reasoning. Extensive experiments across multiple benchmarks show that Alphacast generally outperforms representative baselines. Code is available at this repository: https://github.com/echo01-ai/AlphaCast.