Can AI Master Econometrics? Evidence from Econometrics AI Agent on Expert-Level Tasks
Qiang Chen, Tianyang Han, Jin Li, Ye Luo, Zigan Wang, Yuxiao Wu, Xiaowei Zhang, Tuo Zhou
TL;DR
This work introduces MetricsAI, an econometrics-specific AI agent built on MetaGPT to automate complex empirical analysis. It integrates a domain-tailored tool library with a zero-shot learning framework and a memory-enabled, multi-round workflow to perform tasks such as model selection, data processing, and robustness checks end-to-end. Empirical tests on doctoral coursework and published papers show MetricsAI consistently outperforms baseline LLMs and general AI agents in replication, coefficient and standard error accuracy, and code robustness, achieving high directional replication rates (around 90%+) and substantial perfect replication on coursework. The study highlights potential impacts on democratizing econometric analysis, improving reproducibility, and enabling scalable, no-code workflows for teaching and research, with the open-source toolkit available at the referenced repository.
Abstract
Can AI effectively perform complex econometric analysis traditionally requiring human expertise? This paper evaluates AI agents' capability to master econometrics, focusing on empirical analysis performance. We develop ``MetricsAI'', an Econometrics AI Agent built on the open-source MetaGPT framework. This agent exhibits outstanding performance in: (1) planning econometric tasks strategically, (2) generating and executing code, (3) employing error-based reflection for improved robustness, and (4) allowing iterative refinement through multi-round conversations. We construct two datasets from academic coursework materials and published research papers to evaluate performance against real-world challenges. Comparative testing shows our domain-specialized AI agent significantly outperforms both benchmark large language models (LLMs) and general-purpose AI agents. This work establishes a testbed for exploring AI's impact on social science research and enables cost-effective integration of domain expertise, making advanced econometric methods accessible to users with minimal coding skills. Furthermore, our AI agent enhances research reproducibility and offers promising pedagogical applications for econometrics teaching.