search for




 

애널리스트 보고서 텍스트의 주가예측력에 대한 검증
Verification on stock return predictability of text in analyst reports
Korean J Appl Stat 2023;36(5):489-499
Published online October 31, 2023
© 2023 The Korean Statistical Society.

이영선a, 야마다 아키히코b, 양철원c, 노호석1,a,d
Young-Sun Leea, Akihiko Yamadab, Cheol-Won Yangc, Hohsuk Noh1,a,d

a숙명여자대학교 통계학과; b서울대학교 빅데이터 혁신융합대학; c단국대학교 경영학부; d숙명여자대학교 자연과학연구소

aDepartment of Statistics, Sookmyung Women’s Univesity;
bBigdata Convergence and Open Sharing System, Seoul National Univesity;
cSchool of Business Administration, Dankook Univerisity;
dResearch Institute of Natural Sciences, Sookmyung Women’s Univesity
1Department of Statistics, Sookmyung Women’s Univesity, Cheongpa-ro 47-gil 100, Yongsan-gu, Seoul 04310, Korea. E-mail: hsnoh@sookmyung.ac.kr
Received April 7, 2023; Revised May 1, 2023; Accepted May 13, 2023.
Abstract
온라인 플랫폼을 통한 애널리스트 보고서의 공유가 가능해짐에 따라 애널리스트들이 생성한 보고서는 시장 참여자들 간 금융 정보 격차를 줄일 수 있는 유용한 도구가 되었으며, 애널리스트 보고서의 정량적 정보가 주식수익률 예측에 다수 활용되었다. 하지만 상대적으로 애널리스트 보고서 내 텍스트 정보의 주식수익률 예측 정보력에 대한 국내 자료 기반 연구는 상대적으로 많이 부족하다. 본 연구는 애널리스트 보고서에서 추출 가능한 텍스트로부터 어조 변수를 생성하여 주식수익률 예측에 정보력이 있는지를 검증하되, 기존 연구들의 선형모형 가정 기반 검정의 한계를 해결하고자 랜덤 포레스트 기반의 F-test를 사용하여 기업수익률 예측력을 검증하였다.
As sharing of analyst reports became widely available, reports generated by analysts have become a useful tool to reduce difference in financial information between market participants. The quantitative information of analyst reports has been used in many ways to predict stock returns. However, there are relatively few domestic studies on the prediction power of text information in analyst reports to predict stock returns. We test stock return predictability of text in analyst reports by creating variables representing the TONE from the text. To overcome the limitation of the linear-model-assumption-based approach, we use the random-forest-based F-test.
주요어 : 주식수익률 예측가능성, 자연어 처리, 애널리스트 보고서, 랜덤 포레스트 F-test
Keywords : stock return predictability, natural language processing, analyst reports, random forest F-test
References
  1. Barber BM, Lehavy R, and Trueman B (2010). Ratings changes, ratings levels, and the predictive value of analysts’ recommendations, Financial Management, 39, 533-553.
    CrossRef
  2. Bradley D, Clarke J, Lee S, and Ornthanalai C (2014). Are analysts’ recommendations informative? intraday evidence on the impact of time stamp delays, Journal of Finance, 69, 645-673.
    CrossRef
  3. Cho SS, Byun JH, and Park SH (2012). Short-Selling behavior of investor groups before analyst downgrades, The Korean Journal of Financial Management, 29, 191-231.
  4. Coleman T, PengW, and Mentch L (2022). Scalable and efficient hypothesis testing with random forests, Journal of Machine Learning Research, 23, 1-35.
  5. Davidson R and MacKinnon JG (1981). Several tests for model specification in the presence of alternative hypotheses, Econometrica, 49, 781-793.
    CrossRef
  6. Devlin J, Chang MW, Lee K, and Toutanova K (2019). BERT: Pre-training of deep bidirectional transformers for language understanding. In Proceedings of the 2019 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies, Volume 1 (Long and Short Papers), Minneapolis, Minnesota, 4171-4186,
  7. Huang AH, Zang AY, and Zheng R (2014). Evidence on the information content of text in analyst reports, The Accounting Review, 89, 2151-2180.
    CrossRef
  8. Jang JK, Lee KH, and Lee ZK (2016). How the title of investment strategy report affects stock price forecast: Using text mining method, The Korean Journal of Bigdata, 1, 21-34.
  9. Kim DS and Eum SS (2006). The impact of analysts’ revisions in their stock recommendation and target prices on stock prices, Asia-Pacific Journal of Financial Studies, 35, 75-108.
  10. Kim E and Shin H (2022). KR-FinBert: Fine-tuning KR-FinBert for sentiment analysis,
  11. Kim TH and Lee SY (2013). Do the firm’s exposures to SNS affect their stock prices in Korea?, The Korea Society of Management Information Systems, 491-499.
  12. Liang D, Pan Y, Du Q, and Zhu L (2022). The information content of analysts’ textual reports and stock returns: Evidence from China, Finance Research Letters, 46, 102817.
    CrossRef
  13. McAlexander RJ and Mentch L (2020). Predictive inference with random forests: A new perspective on classical analyses, Research & Politics, 7.
    CrossRef
  14. Yang CW (2021). Information content of analyst report title: Focusing on the TONE of text, Korean Journal of Financial Management, 38, 1-38.


April 2024, 37 (2)