Paper: arXiv 2512.15728

Authors: Yuhan Hou, Tianji Rao, Jeremy Tan, Adler Viton, Xiyue Zhang, David Ye, Abhishek Kodi, Sanjana Dulam, Aditya Paul, Yikai Feng

Abstract

The Federal Open Market Committee (FOMC) sets the federal funds rate, shaping monetary policy and the broader economy. We introduce \emph{FedSight AI}, a multi-agent framework that uses large language models (LLMs) to simulate FOMC deliberations and predict policy outcomes. Member agents analyze structured indicators and unstructured inputs such as the Beige Book, debate options, and vote, replicating committee reasoning. A Chain-of-Draft (CoD) extension further improves efficiency and accuracy by enforcing concise multistage reasoning. Evaluated at 2023-2024 meetings, FedSight CoD achieved accuracy of 93.75% and stability of 93.33%, outperforming baselines including MiniFed and Ordinal Random Forest (RF), while offering transparent reasoning aligned with real FOMC communications.

Complexity vs Empirical Score

  • Math Complexity: 2.5/10
  • Empirical Rigor: 6.5/10
  • Quadrant: Street Traders — practical and empirical, lighter on theory

Why this score: The paper introduces a novel multi-agent LLM framework (FedSight AI) that models FOMC rate decisions through deliberative reasoning rather than statistical black boxes. While mathematically light, it demonstrates strong empirical rigor across 16 FOMC meetings, benchmarking effectively against prior LLM and machine learning models.

Research Flowchart

  flowchart TD
    A["Research Goal: Predict FOMC Federal Funds Rate & Simulate Deliberations"] --> B["Data Inputs: Structured Economic Indicators & Unstructured Beige Book"]
    B --> C["Multi-Agent Architecture: LLM Agents Simulate FOMC Members"]
    C --> D["Processing: Chain-of-Draft Reasoning, Debate & Voting"]
    D --> E["Outcomes: 93.75% Accuracy & 93.33% Stability (2023-2024 Meetings)"]
    E --> F["Key Value: Outperforms Baselines with Transparent Policy Reasoning"]

Cited References (13)

  1. The Federal Reserve’s Response to the COVID-19 Contraction: An Initial Appraisal (2021)
  2. Federal Funds Rate Prediction (2005)
  3. Empirical Asset Pricing via Machine Learning (2020) — link
  4. MiniFed : Integrating LLM-based Agentic-Workflow for Simulating FOMC Meeting (2024) — link
  5. Chain-of-Thought Prompting Elicits Reasoning in Large Language Models (2023) — link
  6. Forecasting the direction of the Fed’s monetary policy decisions using random forest (2024)
  7. Innovations in Interest Rate Policy (1966)
  8. Discretion versus policy rules in practice (1993) — link
and 5 more
  1. Term premia and interest rate forecasts in affine models (2002)
  2. Nowcasting: The real-time informational content of macroeconomic data (2008)
  3. Deep learning for time series classification: a review (2019)
  4. Multi-Agent Collaboration Mechanisms: A Survey of LLMs (2025) — link
  5. Self-Evaluation Improves Selective Generation in Large Language Models (2023) — link

0 of 13 cited works are in the corpus; resolved via arXiv source bundles, Semantic Scholar and OpenAlex. Corpus-wide: 2518 linked papers.