Paper: arXiv 2502.17011

Abstract

Financial bond yield forecasting is challenging due to data scarcity, nonlinear macroeconomic dependencies, and evolving market conditions. In this paper, we propose a novel framework that leverages Causal Generative Adversarial Networks (CausalGANs) and Soft Actor-Critic (SAC) reinforcement learning (RL) to generate high-fidelity synthetic bond yield data for four major bond categories (AAA, BAA, US10Y, Junk). By incorporating 12 key macroeconomic variables, we ensure statistical fidelity by preserving essential market properties. To transform this market dependent synthetic data into actionable insights, we employ a finetuned Large Language Model (LLM) Qwen2.5-7B that generates trading signals (BUY/HOLD/SELL), risk assessments, and volatility projections. We use automated, human and LLM evaluations, all of which demonstrate that our framework improves forecasting performance over existing methods, with statistical validation via predictive accuracy, MAE evaluation(0.103%), profit/loss evaluation (60% profit rate), LLM evaluation (3.37/5) and expert assessments scoring 4.67 out of 5. The reinforcement learning-enhanced synthetic data generation achieves the least Mean Absolute Error of 0.103, demonstrating its effectiveness in replicating real-world bond market dynamics. We not only enhance data-driven trading strategies but also provides a scalable, high-fidelity synthetic financial data pipeline for risk & volatility management and investment decision-making. This work establishes a bridge between synthetic data generation, LLM driven financial forecasting, and language model evaluation, contributing to AI-driven financial decision-making.

Complexity vs Empirical Score

  • Math Complexity: 7.5/10
  • Empirical Rigor: 8.0/10
  • Quadrant: Holy Grail — high math complexity, high empirical rigor

Why this score: The paper employs advanced mathematics including Causal GANs and Soft Actor-Critic RL, but provides extensive empirical validation with specific metrics like MAE (0.103%), profit rates (60%), LLM evaluation scores, and a clear experimental pipeline, indicating high implementation readiness.

Research Flowchart

  flowchart TD
  A["Research Goal: Predict Liquidity-Aware Bond Yields via CausalGAN & DRL"] --> B["Data Input: 4 Bond Categories & 12 Macroeconomic Variables"]
  B --> C["Synthetic Data Generation: CausalGAN & Soft Actor-Critic SAC"]
  C --> D["LLM Processing: Qwen2.5-7B Finetuning for Signals/Risk/Volatility"]
  D --> E["Automated & Human Evaluation"]
  E --> F["Key Outcomes: 0.103% MAE, 60% Profit, 4.67/5 Expert Score"]

Cited References (68)

  1. Attention-Based LSTM for Bond Yield Forecasting (2023)
  2. Using Barbells to Lift Risk-Adjusted Return (2020)
  3. Riding the yield curve: a variety of strategies (2005)
  4. Portfolio Optimization using Deep Reinforcement Learning (2024)
  5. Adaptive learning by genetic algorithms: Analytical results and applications to economic models (2011)
  6. The FRB/US Model: A Tool for Macroeconomic Policy Analysis (2013)
  7. DeepSeek R1: A Large Language Model for Robust Decision Evaluation (2024) — link
  8. Regulatory Constrained GANs for Fixed-Income Markets (2023)
and 60 more
  1. Qwen2. 5 technical report (2024) — link
  2. Statistical Validation of Synthetic Time Series (2023)
  3. RAG for Low-Data Financial Forecasting (2023)
  4. Time-series Generative Adversarial Networks (2019) — link
  5. Soft Actor-Critic: Off-Policy Maximum Entropy Deep Reinforcement Learning with a Stochastic Actor (2018)
  6. Generative Model for Financial Time Series Trained with MMD Using a Signature Kernel — → 2407.19848
  7. Deep Learning Based on GANs and CNNs for Financial Time Series Predictions (2020) — link
  8. Time-FFM: Towards LM-Empowered Federated Foundation Model for Time Series Forecasting (2024)
  9. TEMPO: Prompt-based Generative Pre-trained Transformer for Time Series Forecasting (2024)
  10. GPT4MTS: Prompt-based Large Language Model for Multimodal Time-series Forecasting (2024)
  11. MOMENT: A Family of Open Time-series Foundation Models (2024)
  12. TIME-LLM: Time Series Forecasting by Reprogramming Large Language Models (2024)
  13. Timer: Generative Pre-trained Transformers Are Large Time Series Models (2024)
  14. Unified Training of Universal Time Series Forecasting Transformers (2024)
  15. TradingAgents: Multi-Agents LLM Financial Trading Framework (2025)
  16. FinCon: A Synthesized LLM Multi-Agent System for Enhanced Financial Decision Making (2024)
  17. Term-Structure Models: A Graduate Course (2009)
  18. Portfolio Optimization Using Deep Reinforcement Learning (2021)
  19. FinBERT: Financial Sentiment Analysis with Pre-trained Language Models (2019) — link
  20. ‘Finance Wizard’ at the FinLLM Challenge Task: Financial Text Summarization (2024) — link
  21. Generative Adversarial Nets (2014)
  22. Wasserstein GAN (2017)
  23. Human-level control through deep reinforcement learning (2015)
  24. Deep Learning with Long Short-Term Memory Networks for Financial Market Predictions (2018)
  25. Forecasting the term structures of Treasury and corporate yields using dynamic Nelson-Siegel models (2011)
  26. FEB-stacking and FEB-DNN models for stock trend prediction: a performance analysis for pre and post covid-19 periods (2021)
  27. Forecasting government bond yields with large Bayesian vector autoregressions (2012)
  28. HuggingFace’s Transformers: State-of-the-art Natural Language Processing (2020) — link
  29. PyTorch: An Imperative Style, High-Performance Deep Learning Library (2019) — link
  30. Integrating Stock Features and Global Information via Large Language Models for Enhanced Stock Return Prediction (2023) — link
  31. PIXIU: A Large Language Model, Instruction Data and Evaluation Benchmark for Finance (2023) — link
  32. LLMFactor: Extracting Profitable Factors through Prompts for Explainable Stock Movement Prediction (2024) — link
  33. Can ChatGPT Forecast Stock Price Movements? Return Predictability and Large Language Models (2024) — link
  34. Can Large Language Models beat wall street? Evaluating GPT-4’s impact on financial decision-making with MarketSenseAI (2024) — link
  35. Fine-Tuning Large Language Models for Stock Return Prediction Using Newsflow (2024) — link
  36. Financial Statement Analysis with Large Language Models (2024) — link
  37. Sentiment trading with large language models — → 2412.19245
  38. Using generative adversarial networks to synthesize artificial financial datasets (2020) — link
  39. Quant GANs: Deep Generation of Financial Time Series (2019) — link
  40. Generating Realistic Stock Market Order Streams (2020) — link
  41. Generative Adversarial Networks Applied to Synthetic Financial Scenarios Generation (2022) — link
  42. Market-GAN: Adding Control to Financial Market Data Generation with Semantic Context — → 2309.07708
  43. Simulating financial time series using attention (2022) — link
  44. Deep Hedging: Learning to Simulate Equity Option Markets (2019) — link
  45. Deep Reinforcement Learning for Trading (2019) — link
  46. Optimal Market Making by Reinforcement Learning (2021) — link
  47. Financial Trading as a Game: A Deep Reinforcement Learning Approach (2018) — link
  48. Model-based Deep Reinforcement Learning for Dynamic Portfolio Optimization (2019) — link
  49. FinRL-Podracer: High Performance and Scalable Deep Reinforcement Learning for Quantitative Finance (2021) — link
  50. An Application of Deep Reinforcement Learning to Algorithmic Trading (2020) — link
  51. The Evolution of Reinforcement Learning in Quantitative Finance: A Survey (2024) — link
  52. Deep reinforcement learning for portfolio management (2022) — link
  53. Recent advances in reinforcement learning in finance (2023)
  54. SuperCLUE-Fin: Graded Fine-Grained Analysis of Chinese LLMs on Diverse Financial Tasks and Applications (2024) — link
  55. Empowering Many, Biasing a Few: Generalist Credit Scoring through Large Language Models (2024) — link
  56. Evaluating Financial Relational Graphs: Interpretation Before Prediction (2024) — link
  57. FinBen: A Holistic Financial Benchmark for Large Language Models (2024) — link
  58. LOB-Bench: Benchmarking Generative AI for Finance – an Application to Limit Order Book Data (2025) — link
  59. Qwen/Qwen2.5-7B-Instruct-1M · Hugging Face (2025) — link
  60. deepseek-ai/DeepSeek-R1-Distill-Qwen-32B · Hugging Face (2025) — link

3 of 68 cited works are in the corpus; resolved via arXiv source bundles, Semantic Scholar and OpenAlex. Corpus-wide: 2518 linked papers.