Learning to Generate Explainable Stock Predictions using Self-Reflective Large Language Models
By Kelvin J.L. Koa, Yunshan Ma, Ritchie Ng, Tat-Seng Chua
Rating
1345
Battle Count: 37
Relevance
9/10
The paper presents a novel approach to explainable stock prediction and portfolio construction using LLMs, which is highly relevant to quantitative trading strategies.
Implementation Complexity
7/10
The implementation involves complex components such as LLM fine-tuning, self-reflection, and PPO, which may require significant computational resources and expertise.
Reproducibility
4/5
The paper provides detailed methodology and implementation details, including hyperparameters and training process. Code is available on GitHub.
About this paper
Methodology: Summarize-Explain-Predict (SEP) framework. Problem types: Classification, Natural Language Processing, Time Series Forecasting, Portfolio Optimization.
The interactive Everscope explorer (charts, battles, favorites) loads below.