情绪在小盘股中起作用:日本股市的情绪与语言模型

Masahiro Suzuki , Yasushi Ishikawa , Masayuki Teraguchi , Hiroki Sakaji
{"title":"情绪在小盘股中起作用:日本股市的情绪与语言模型","authors":"Masahiro Suzuki ,&nbsp;Yasushi Ishikawa ,&nbsp;Masayuki Teraguchi ,&nbsp;Hiroki Sakaji","doi":"10.1016/j.jjimei.2024.100318","DOIUrl":null,"url":null,"abstract":"<div><div>We calculate sentiment from the Japanese Company Handbook, which contains a compact overview of Japanese companies’ business situation and financial data, using multiple methods, including large language models. Language models such as BERT and ChatGPT are advancing the application of natural language processing (NLP) to financial fields. We construct multiple sentiment calculation methods using sentiment dictionaries, models trained on existing sentiment datasets, ChatGPT, and GPT-4. Our analysis shows that stocks with higher sentiment scores tend to have higher excess returns, while those with lower scores tend to have lower excess returns. This feature is enhanced particularly in small-cap stocks. Comparisons between the models showed higher returns at high sentiment for the model trained with the existing sentiment dataset and lower returns at low sentiment for ChatGPT. The DeBERTaV2 model trained on Economy Watchers Survey data performed best in terms of returns at the highest sentiment quantile.</div></div>","PeriodicalId":100699,"journal":{"name":"International Journal of Information Management Data Insights","volume":"5 1","pages":"Article 100318"},"PeriodicalIF":0.0000,"publicationDate":"2025-01-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":"{\"title\":\"Sentiment works in small-cap stocks: Japanese stock’s sentiment with language models\",\"authors\":\"Masahiro Suzuki ,&nbsp;Yasushi Ishikawa ,&nbsp;Masayuki Teraguchi ,&nbsp;Hiroki Sakaji\",\"doi\":\"10.1016/j.jjimei.2024.100318\",\"DOIUrl\":null,\"url\":null,\"abstract\":\"<div><div>We calculate sentiment from the Japanese Company Handbook, which contains a compact overview of Japanese companies’ business situation and financial data, using multiple methods, including large language models. Language models such as BERT and ChatGPT are advancing the application of natural language processing (NLP) to financial fields. We construct multiple sentiment calculation methods using sentiment dictionaries, models trained on existing sentiment datasets, ChatGPT, and GPT-4. Our analysis shows that stocks with higher sentiment scores tend to have higher excess returns, while those with lower scores tend to have lower excess returns. This feature is enhanced particularly in small-cap stocks. Comparisons between the models showed higher returns at high sentiment for the model trained with the existing sentiment dataset and lower returns at low sentiment for ChatGPT. The DeBERTaV2 model trained on Economy Watchers Survey data performed best in terms of returns at the highest sentiment quantile.</div></div>\",\"PeriodicalId\":100699,\"journal\":{\"name\":\"International Journal of Information Management Data Insights\",\"volume\":\"5 1\",\"pages\":\"Article 100318\"},\"PeriodicalIF\":0.0000,\"publicationDate\":\"2025-01-15\",\"publicationTypes\":\"Journal Article\",\"fieldsOfStudy\":null,\"isOpenAccess\":false,\"openAccessPdf\":\"\",\"citationCount\":\"0\",\"resultStr\":null,\"platform\":\"Semanticscholar\",\"paperid\":null,\"PeriodicalName\":\"International Journal of Information Management Data Insights\",\"FirstCategoryId\":\"1085\",\"ListUrlMain\":\"https://www.sciencedirect.com/science/article/pii/S2667096824001071\",\"RegionNum\":0,\"RegionCategory\":null,\"ArticlePicture\":[],\"TitleCN\":null,\"AbstractTextCN\":null,\"PMCID\":null,\"EPubDate\":\"\",\"PubModel\":\"\",\"JCR\":\"\",\"JCRName\":\"\",\"Score\":null,\"Total\":0}","platform":"Semanticscholar","paperid":null,"PeriodicalName":"International Journal of Information Management Data Insights","FirstCategoryId":"1085","ListUrlMain":"https://www.sciencedirect.com/science/article/pii/S2667096824001071","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"","JCRName":"","Score":null,"Total":0}
引用次数: 0

摘要

我们从日本公司手册中计算情绪,该手册包含日本公司业务状况和财务数据的简明概述,使用多种方法,包括大型语言模型。BERT和ChatGPT等语言模型正在推动自然语言处理(NLP)在金融领域的应用。我们使用情感词典、基于现有情感数据集训练的模型、ChatGPT和GPT-4构建了多种情感计算方法。我们的分析表明,情绪得分较高的股票往往有较高的超额回报,而得分较低的股票往往有较低的超额回报。这一特点在小盘股中尤为突出。模型之间的比较表明,使用现有情绪数据集训练的模型在高情绪下的回报较高,而ChatGPT在低情绪下的回报较低。根据经济观察调查(economic Watchers Survey)数据训练的DeBERTaV2模型在情绪最高分位数的回报率方面表现最好。
本文章由计算机程序翻译,如有差异,请以英文原文为准。
Sentiment works in small-cap stocks: Japanese stock’s sentiment with language models
We calculate sentiment from the Japanese Company Handbook, which contains a compact overview of Japanese companies’ business situation and financial data, using multiple methods, including large language models. Language models such as BERT and ChatGPT are advancing the application of natural language processing (NLP) to financial fields. We construct multiple sentiment calculation methods using sentiment dictionaries, models trained on existing sentiment datasets, ChatGPT, and GPT-4. Our analysis shows that stocks with higher sentiment scores tend to have higher excess returns, while those with lower scores tend to have lower excess returns. This feature is enhanced particularly in small-cap stocks. Comparisons between the models showed higher returns at high sentiment for the model trained with the existing sentiment dataset and lower returns at low sentiment for ChatGPT. The DeBERTaV2 model trained on Economy Watchers Survey data performed best in terms of returns at the highest sentiment quantile.
求助全文
通过发布文献求助,成功后即可免费获取论文全文。 去求助
来源期刊
CiteScore
19.20
自引率
0.00%
发文量
0
×
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
确定
请完成安全验证×
copy
已复制链接
快去分享给好友吧!
我知道了
右上角分享
点击右上角分享
0
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 布克学术 All rights reserved.
京ICP备2023020795号-1
ghs 京公网安备 11010802042870号
Book学术文献互助
Book学术文献互助群
群 号:481959085
Book学术官方微信