Harnessing Earnings Reports for Stock Predictions: A QLoRA-Enhanced LLM Approach

Haowei Ni, Shuchen Meng, Xupeng Chen, Ziqing Zhao, Andi Chen, Panfeng Li, Shiyao Zhang, Qifu Yin, Yuanqing Wang, Yuxi Chan
{"title":"Harnessing Earnings Reports for Stock Predictions: A QLoRA-Enhanced LLM Approach","authors":"Haowei Ni, Shuchen Meng, Xupeng Chen, Ziqing Zhao, Andi Chen, Panfeng Li, Shiyao Zhang, Qifu Yin, Yuanqing Wang, Yuxi Chan","doi":"arxiv-2408.06634","DOIUrl":null,"url":null,"abstract":"Accurate stock market predictions following earnings reports are crucial for\ninvestors. Traditional methods, particularly classical machine learning models,\nstruggle with these predictions because they cannot effectively process and\ninterpret extensive textual data contained in earnings reports and often\noverlook nuances that influence market movements. This paper introduces an\nadvanced approach by employing Large Language Models (LLMs) instruction\nfine-tuned with a novel combination of instruction-based techniques and\nquantized low-rank adaptation (QLoRA) compression. Our methodology integrates\n'base factors', such as financial metric growth and earnings transcripts, with\n'external factors', including recent market indices performances and analyst\ngrades, to create a rich, supervised dataset. This comprehensive dataset\nenables our models to achieve superior predictive performance in terms of\naccuracy, weighted F1, and Matthews correlation coefficient (MCC), especially\nevident in the comparison with benchmarks such as GPT-4. We specifically\nhighlight the efficacy of the llama-3-8b-Instruct-4bit model, which showcases\nsignificant improvements over baseline models. The paper also discusses the\npotential of expanding the output capabilities to include a 'Hold' option and\nextending the prediction horizon, aiming to accommodate various investment\nstyles and time frames. This study not only demonstrates the power of\nintegrating cutting-edge AI with fine-tuned financial data but also paves the\nway for future research in enhancing AI-driven financial analysis tools.","PeriodicalId":501139,"journal":{"name":"arXiv - QuantFin - Statistical Finance","volume":"42 1","pages":""},"PeriodicalIF":0.0000,"publicationDate":"2024-08-13","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"arXiv - QuantFin - Statistical Finance","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/arxiv-2408.06634","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"","JCRName":"","Score":null,"Total":0}
引用次数: 0

Abstract

Accurate stock market predictions following earnings reports are crucial for investors. Traditional methods, particularly classical machine learning models, struggle with these predictions because they cannot effectively process and interpret extensive textual data contained in earnings reports and often overlook nuances that influence market movements. This paper introduces an advanced approach by employing Large Language Models (LLMs) instruction fine-tuned with a novel combination of instruction-based techniques and quantized low-rank adaptation (QLoRA) compression. Our methodology integrates 'base factors', such as financial metric growth and earnings transcripts, with 'external factors', including recent market indices performances and analyst grades, to create a rich, supervised dataset. This comprehensive dataset enables our models to achieve superior predictive performance in terms of accuracy, weighted F1, and Matthews correlation coefficient (MCC), especially evident in the comparison with benchmarks such as GPT-4. We specifically highlight the efficacy of the llama-3-8b-Instruct-4bit model, which showcases significant improvements over baseline models. The paper also discusses the potential of expanding the output capabilities to include a 'Hold' option and extending the prediction horizon, aiming to accommodate various investment styles and time frames. This study not only demonstrates the power of integrating cutting-edge AI with fine-tuned financial data but also paves the way for future research in enhancing AI-driven financial analysis tools.
查看原文
分享 分享
微信好友 朋友圈 QQ好友 复制链接
本刊更多论文
利用盈利报告进行股票预测:QLoRA 增强型 LLM 方法
准确预测财报发布后的股市行情对投资者至关重要。传统方法,尤其是经典的机器学习模型,在这些预测方面举步维艰,因为它们无法有效处理和解释财报中包含的大量文本数据,而且经常忽略影响市场走势的细微差别。本文介绍了一种先进的方法,即采用基于指令的技术和量化低秩适应(QLoRA)压缩的新组合对大语言模型(LLMs)进行指令微调。我们的方法将财务指标增长和盈利记录等 "基础因素 "与近期市场指数表现和分析师评级等 "外部因素 "整合在一起,创建了一个丰富的监督数据集。这种全面的数据集使我们的模型在准确性、加权 F1 和马修斯相关系数 (MCC) 方面取得了卓越的预测性能,这在与 GPT-4 等基准的比较中尤为明显。我们特别强调了llama-3-8b-Instruct-4bit 模型的功效,它比基准模型有了显著的改进。本文还讨论了扩展输出功能的潜力,包括 "持有 "选项和延长预测期限,旨在适应各种投资风格和时间框架。这项研究不仅展示了尖端人工智能与微调金融数据相结合的威力,还为未来加强人工智能驱动的金融分析工具的研究铺平了道路。
本文章由计算机程序翻译,如有差异,请以英文原文为准。
求助全文
约1分钟内获得全文 去求助
来源期刊
自引率
0.00%
发文量
0
期刊最新文献
Macroscopic properties of equity markets: stylized facts and portfolio performance Tuning into Climate Risks: Extracting Innovation from TV News for Clean Energy Firms On the macroeconomic fundamentals of long-term volatilities and dynamic correlations in COMEX copper futures Market information of the fractional stochastic regularity model Critical Dynamics of Random Surfaces
×
引用
GB/T 7714-2015
复制
MLA
复制
APA
复制
导出至
BibTeX EndNote RefMan NoteFirst NoteExpress
×
×
提示
您的信息不完整,为了账户安全,请先补充。
现在去补充
×
提示
您因"违规操作"
具体请查看互助需知
我知道了
×
提示
现在去查看 取消
×
提示
确定
0
微信
客服QQ
Book学术公众号 扫码关注我们
反馈
×
意见反馈
请填写您的意见或建议
请填写您的手机或邮箱
已复制链接
已复制链接
快去分享给好友吧!
我知道了
×
扫码分享
扫码分享
Book学术官方微信
Book学术文献互助
Book学术文献互助群
群 号:481959085
Book学术
文献互助 智能选刊 最新文献 互助须知 联系我们:info@booksci.cn
Book学术提供免费学术资源搜索服务,方便国内外学者检索中英文文献。致力于提供最便捷和优质的服务体验。
Copyright © 2023 Book学术 All rights reserved.
ghs 京公网安备 11010802042870号 京ICP备2023020795号-1