TimeSeriesScientist: A General-Purpose AI Agent for Time Series Analysis

2025-10-01

AI Summary

Time-series forecasting is critical for decision-making across domains, yet real-world applications face challenges including short sequence lengths, high noise levels, and strong heterogeneity—rendering manual preprocessing and model tuning more costly than modeling itself. To address this, we propose the first large language model (LLM)-driven multi-agent framework for general-purpose time-series analysis. It comprises four specialized agents—Curator, Planner, Forecaster, and Reporter—that jointly perform multimodal diagnostic analysis, adaptive model selection, ensemble strategy optimization, and natural-language report generation. The framework employs a white-box reasoning mechanism with external tool invocation and closed-loop validation, substantially reducing human intervention. Evaluated on eight benchmark datasets, it achieves average prediction errors 10.4% lower than traditional statistical models and 38.2% lower than LLM-based baselines, while enhancing interpretability and cross-task generalization.

Technology Category

• Machine Learning: Time-Series/Data Streams
• Natural Language Processing: (Large) Language Models
• Planning, Routing, and Scheduling: Planning with Language Models

Application Category

• User Modeling, Personalization and Recommendation: Large Language Models (LLM) for user modeling and recommendation
• Semantics and Knowledge: Data modeling to support human-machine intelligence, including LLMs agents, intelligent system behavior, explanations, and user-friendly interactions
• Search and Retrieval-Augmented AI: Search Tool Learning with LLM: Teaching LLMs to invoke search and make use of retrieved information

Abstract

Time series forecasting is central to decision-making in domains as diverse as energy, finance, climate, and public health. In practice, forecasters face thousands of short, noisy series that vary in frequency, quality, and horizon, where the dominant cost lies not in model fitting, but in the labor-intensive preprocessing, validation, and ensembling required to obtain reliable predictions. Prevailing statistical and deep learning models are tailored to specific datasets or domains and generalize poorly. A general, domain-agnostic framework that minimizes human intervention is urgently in demand. In this paper, we introduce TimeSeriesScientist (TSci), the first LLM-driven agentic framework for general time series forecasting. The framework comprises four specialized agents: Curator performs LLM-guided diagnostics augmented by external tools that reason over data statistics to choose targeted preprocessing; Planner narrows the hypothesis space of model choice by leveraging multi-modal diagnostics and self-planning over the input; Forecaster performs model fitting and validation and, based on the results, adaptively selects the best model configuration as well as ensemble strategy to make final predictions; and Reporter synthesizes the whole process into a comprehensive, transparent report. With transparent natural-language rationales and comprehensive reports, TSci transforms the forecasting workflow into a white-box system that is both interpretable and extensible across tasks. Empirical results on eight established benchmarks demonstrate that TSci consistently outperforms both statistical and LLM-based baselines, reducing forecast error by an average of 10.4% and 38.2%, respectively. Moreover, TSci produces a clear and rigorous report that makes the forecasting workflow more transparent and interpretable.

Problem

• Automating labor-intensive preprocessing and validation for noisy time series data
• Overcoming poor generalization of domain-specific forecasting models
• Creating interpretable forecasting workflows with minimal human intervention

Innovation

• LLM-driven multi-agent framework for time series
• Automated preprocessing and model selection via agents
• Generates interpretable reports with transparent rationales