Predicting Liquidity-Aware Bond Yields using Causal GANs and Deep Reinforcement Learning with LLM Evaluation

arXiv cs.CL / 4/27/2026

💬 OpinionDeveloper Stack & InfrastructureTools & Practical UsageModels & Research

共有:

Key Points

The paper addresses bond yield forecasting difficulties caused by scarce data, nonlinear macroeconomic relationships, and changing market conditions by proposing a new framework.
It combines Causal Generative Adversarial Networks (CausalGANs) with Soft Actor-Critic (SAC) reinforcement learning to generate high-fidelity synthetic yield data for four bond categories (AAA, BAA, US10Y, and Junk) while preserving key market statistical properties using 12 macroeconomic variables.
To make the synthetic market-dependent data actionable, the authors fine-tune a Qwen2.5-7B large language model to output trading signals (BUY/HOLD/SELL), risk assessments, and volatility projections.
The framework is evaluated with automated, human, and LLM-based assessments, reporting improved forecasting performance using metrics such as MAE (0.103%), profit/loss outcomes (60% profit rate), and scoring of 3.37/5 (LLM) and 4.67/5 (experts).
Overall, the work claims a scalable synthetic financial data pipeline that bridges causal synthetic-data generation, LLM-driven forecasting, and language-model evaluation for risk and volatility management.

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.