[논문]증권신고서의 TF-IDF 텍스트 분석과 기계학습을 이용한 공모주의 상장 이후 주가 등락 예측

양수연; 이채록; 원종관; 홍태호

doi:10.13088/jiis.2022.28.2.237

증권신고서의 TF-IDF 텍스트 분석과 기계학습을 이용한 공모주의 상장 이후 주가 등락 예측
The prediction of the stock price movement after IPO using machine learning and text analysis based on TF-IDF 원문보기

지능정보연구 = Journal of intelligence and information systems, v.28 no.2, 2022년, pp.237 - 262

양수연 (KAIST 경영대학원 경영공학부) , 이채록 (부산대학교 경영학과) , 원종관 (부산대학교 경영학과) , 홍태호 (부산대학교 경영학과)

초록
AI-Helper

본 연구는 개인투자자들의 투자의사결정에 도움을 주고자, 증권신고서의 TF-IDF 텍스트 분석과 기계학습을 이용해 공모주의 상장 5거래일 이후 주식 가격 등락을 예측하는 모델을 제시한다. 연구 표본은 2009년 6월부터 2020년 12월 사이에 신규 상장된 691개의 국내 IPO 종목이다. 기업, 공모, 시장과 관련된 다양한 재무적 및 비재무적 IPO 관련 변수와 증권신고서의 어조를 분석하여 예측했고, 증권신고서의 어조 분석을 위해서 TF-IDF (Term Frequency - Inverse Document Frequency)에 기반한 텍스트 분석을 이용해 신고서의 투자위험요소란의 텍스트를 긍정적 어조, 중립적 어조, 부정적 어조로 분류하였다. 가격 등락 예측에는 로지스틱 회귀분석(Logistic Regression), 랜덤 포레스트(Random Forest), 서포트벡터머신(Support Vector Machine), 인공신경망(Artificial Neural Network) 기법을 사용하였고, 예측 결과 IPO 관련 변수와 증권신고서 어조 변수를 함께 사용한 모델이 IPO 관련 변수만을 사용한 모델보다 높은 예측 정확도를 보였다. 랜덤 포레스트 모형은 1.45%p 높아진 예측 정확도를 보였으며, 인공신공망 모형과 서포트벡터머신 모형은 각각 4.34%p, 5.07%p 향상을 보였다. 추가적으로 모형간 차이를 맥니마 검정을 통해 통계적으로 검증한 결과, 어조 변수의 유무에 따른 예측 모형의 성과 차이가 유의확률 1% 수준에서 유의했다. 이를 통해, 증권신고서에 표현된 어조가 공모주의 가격 등락 예측에 영향을 미치는 요인이라는 것을 확인할 수 있었다.

Abstract ▼ AI-Helper

There has been a growing interest in IPOs (Initial Public Offerings) due to the profitable returns that IPO stocks can offer to investors. However, IPOs can be speculative investments that may involve substantial risk as well because shares tend to be volatile, and the supply of IPO shares is often highly limited. Therefore, it is crucially important that IPO investors are well informed of the issuing firms and the market before deciding whether to invest or not. Unlike institutional investors, individual investors are at a disadvantage since there are few opportunities for individuals to obtain information on the IPOs. In this regard, the purpose of this study is to provide individual investors with the information they may consider when making an IPO investment decision. This study presents a model that uses machine learning and text analysis to predict whether an IPO stock price would move up or down after the first 5 trading days. Our sample includes 691 Korean IPOs from June 2009 to December 2020. The input variables for the prediction are three tone variables created from IPO prospectuses and quantitative variables that are either firm-specific, issue-specific, or market-specific. The three prospectus tone variables indicate the percentage of positive, neutral, and negative sentences in a prospectus, respectively. We considered only the sentences in the Risk Factors section of a prospectus for the tone analysis in this study. All sentences were classified into 'positive', 'neutral', and 'negative' via text analysis using TF-IDF (Term Frequency - Inverse Document Frequency). Measuring the tone of each sentence was conducted by machine learning instead of a lexicon-based approach due to the lack of sentiment dictionaries suitable for Korean text analysis in the context of finance. For this reason, the training set was created by randomly selecting 10% of the sentences from each prospectus, and the sentence classification task on the training set was performed after reading each sentence in person. Then, based on the training set, a Support Vector Machine model was utilized to predict the tone of sentences in the test set. Finally, the machine learning model calculated the percentages of positive, neutral, and negative sentences in each prospectus. To predict the price movement of an IPO stock, four different machine learning techniques were applied: Logistic Regression, Random Forest, Support Vector Machine, and Artificial Neural Network. According to the results, models that use quantitative variables using technical analysis and prospectus tone variables together show higher accuracy than models that use only quantitative variables. More specifically, the prediction accuracy was improved by 1.45% points in the Random Forest model, 4.34% points in the Artificial Neural Network model, and 5.07% points in the Support Vector Machine model. After testing the performance of these machine learning techniques, the Artificial Neural Network model using both quantitative variables and prospectus tone variables was the model with the highest prediction accuracy rate, which was 61.59%. The results indicate that the tone of a prospectus is a significant factor in predicting the price movement of an IPO stock. In addition, the McNemar test was used to verify the statistically significant difference between the models. The model using only quantitative variables and the model using both the quantitative variables and the prospectus tone variables were compared, and it was confirmed that the predictive performance improved significantly at a 1% significance level.

주제어

표/그림 (15)

표 Studies on Predicting the IPO Price Using Machine Learning
그림 Research Framework
표 Distribution of Sample IPOs by Year
표 Examples of Positive/Neutral/Negative Sentences from Prospectuses
표 Sentence Classification Results
그림 Word Cloud of Positive Sentences
그림 Word Cloud of Neutral Sentences
그림 Word Cloud of Negative Sentences
표 5-Fold Cross-Validation Results
표 Variable Description
표 Descriptive Statistics of Variables
표 IPO Variable Selection Results
표 The Distribution of Stock Price Movement After IPO in Training and Test Set
표 Comparison of Performance Results
표 McNemar Test Results

참고문헌 (77)

Ahn, C. K. and D. Kim, "Efficient variable selection method using conditional mutual information," Journal of the Korean Data and Information Science Society, Vol.25, No.5(2014), 1079~1094.
Ahn, H. C., "Optimization of Multiclass Support Vector Machine using Genetic Algorithm: Application to the Prediction of Corporate Credit Rating," Information Systems Review, Vol.16, No.3(2014), 161~177.
Baba, B. and G. Sevil, "Predicting IPO initial returns using random forest," Borsa Istanbul Review, Vol.20, No.1(2020), 13~23.

상세보기
Baek, J. S. and M. S. Jeong, "A study on the effect of competition rate of subscription and guarantee rate on the underpricing in the initial public offerings," The Korean Finance Association, (2018), 1943~1980.
Basti, E., C. Kuzey, and D. Delen, "Analyzing initial public offerings' short-term performance using decision trees and SVMs," Decision Support Systems, Vol.73, (2015), 15~27.

상세보기
Benveniste, L. M., and P. A. Spindt, "How investment bankers determine the offer price and allocation of new issues," Journal of Financial Economics, Vol.24, No.2(1989), 343~361.

상세보기
Breiman, L, "Random forests," Machine Learning, Vol.45, No. 1(2001), 5~32.
Brown, I. and C. Mues, "An experimental comparison of classification algorithms for imbalanced credit scoring data sets," Expert Systems with Applications, Vol.39, No.3(2012), 3446~3453.

상세보기
Buehlmaier, M. M. M. and T. M. Whited, (2018) "Are Financial Constraints Priced? Evidence from Textual Analysis," The Review of Financial Studies, Vol.31, No.7(2018), 2693~2728.

상세보기
Chan, Y, "Retail Trading and IPO Returns in the Aftermarket," Financial Management, Vol.39, No.4(2010), 1475~1495.
Chen, Y. S. and C. H. Cheng, "A soft-computing based rough sets classifier for classifying IPO returns in the financial markets," Applied Soft Computing, Vol.12, No.1(2012), 462~475.

상세보기
Cho, D. H., H. S. Ryou, S. H. Jung, and K. J. Oh, "Using AI to develop forecasting model in IPO market," Journal of the Korean Data and Information Science Society, Vol.31, No.3 (2020), 579~590.

상세보기
Chun, K. M., I. H. Gee, and H. U. Lee, "The Effect of IPO Subscription Rates for Institutional Investors and Private Investors on IPO Firm Performance: The Moderating Role of Competition and On-line Reviews," Korean Journal of Business Administration, Vol.26, No.5(2013), 1149~1176.

상세보기
Cortes, C. and V. Vapnik, "Support-vector networks," Machine learning, Vol.20, No.3(1995), 273~297.

상세보기
Dadgar, S.M., M.S. Araghi., and M.M. Farahani, "A novel text mining approach based on TF-IDF and Support Vector Machine for news classification" 2016 IEEE International Conference on Engineering and Technology (ICETECH), (2016), 112~116.
Derrien, F, "IPO Pricing in "Hot" Market Conditions: Who Leaves Money on the Table?," The Journal of Finance, Vol.60, No.1(2005), 487~521.

상세보기
Esfahanipour, A., M. Goodarzi., and R. Jahanbin, "Analysis and forecasting of IPO underpricing," Neural Computing and Applications, Vol.27, No.3(2015), 651~658.
Fang, F., K. Dutta, and A. Datta, "Domain Adaptation for Sentiment Classification in Light of Multiple Sources," INFORMS Journal on Computing, Vol.26, No.3(2014), 586-598.

상세보기
Fuksa, M, "Sentiment and the Performance of Initial Public Offerings," Available at SSRN 2243379, (2013).
Gandoman, S. H., N. Kiamehr., and M. Hemetfar, "Forecasting Initial Public Offering Pricing Using Particle Swarm Optimization (PSO) Algorithm and Support Vector Machine (SVM) In Iran," Business and Economic Research, Vol.7, No.1(2017), 336.

상세보기
Garcia, D. "Sentiment during Recessions," The Journal of Finance, Vol.68, No. 3(2013), 1267~1300.

상세보기
Han, G. S. "A Study on the Underpricing of IPOs in Korea Capital Market," Korean International Accounting Review, Vol.59, (2015), 125~146.

상세보기
Hanley, K. W. and G. Hoberg, "The Information Content of IPO Prospectuses," Review of Financial Studies, Vol.23, No.7(2010), 2821~2864.
Hong, S. H. and K. S. Shin, "Using GA based Input Selection Method for Artificial Neural Network Modeling: Application to Bankruptcy Prediction," Journal of Intelligence and Information Systems, Vol.9, No.1(2003), 227~249.
Hong. T. H. and E. M. Kim, "The Prediction of Purchase Amount of Customers Using Support Vector Regression with Separated Learning Method," Journal of Intelligence and Information Systems, Vol.16, No.4(2010), 213~225.
Ibbotson, R. G, "Price performance of common stock new issues," Journal of Financial Economics, Vol.2, No.3(1975), 235~272.

상세보기
Imamah and F. H. Rachman, "Twitter Sentiment Analysis of Covid-19 Using Term Weighting TF-IDF And Logistic Regression," 2020 6th Information Technology International Seminar (ITIS), Surabaya, Indonesia, (2020), 238~242.
Islam, M., F.E. Jubayer, and S.I. Ahmed, "A support vector machine mixed with TF-IDF algorithm to categorize Bengali document," 2017 International Conference on Electrical, Computer and Communication Engineering, (2017), 191~196.
Jegadeesh, N. and D. Wu, "Word power: A new approach for content analysis," Journal of Financial Economics, Vol.110, No.3(2013), 712~729.

상세보기
Joh, S. W. and Y. Kim, "Is Textual Information Informative to Informed Investors? Evidence from Bidding Information of Institutional Investors in IPOs," The Thirteenth Conference on Asia-Pacific Financial Markets, Seoul, Korea, (2018).
Jung, J. Y. and K. W. Park, "A Study on Investor Protection through Morphological Analysis of Equity Crowdfunding Investment Manual," Journal of Information Technology Services, Vol.18, No.5(2019), 165~182.
Kaohua, Y. and Liwen, Z, "Analysis of influencing factors of IPO underpricing based on rough set and support vector machine," 2012 International Conference on Information Management, Innovation Management and Industrial Engineering, Vol. 3, (2012), 244~248.
Katsafados, A. G., I. Androutsopoulos., I. Chalkidis., E. Fergadiotis., G. N. Leledakis., and E. G. Pyrgiotakis, "Textual information and IPO underpricing: A machine learning approach," MPRA Paper 103813, University Library of Munich, Germany, (2020).
Kim, D. Y. and D. E. Won, "An AI Model for Short-term KOSPI Prediction: A Machine Learning-Based Model Using Random Forest Technique," Quantitative Issue, Samsung Securities Research Center, (2019).
Kim, H. A. and S. C. Jung, (2010) "The Effect of Optimistic Investors' Sentiment on Anomalous Behaviors in the Hot Market IPOs," The Korean Journal of Financial Management, Vol.27, No.2(2010), 1~33.
Kim, H. H. "Suggestions for Disclosure of Risk Factors in Investment," Practical Explanations of Securities, Korean Listed Companies Association, (2008), 105~108.
Kim, H. J., J. W. Park, and J. W. Lee, "A Study on the Textual Analysis Research Environment using the DART System in Korea," Korean Accounting Journal, Vol.24, No.4(2015), 199~221.
Kim, I. H., "Ways to Go Public: Choice of IPO, Sellout, and Reverse Takeover," The Korean Finance Association, (2008), 958~1008.
Kim, J., S.M. Jun., S. Hwang., H.K. Kim., J. Heo., and M.S. Kang, "Impact of Activation Functions on Flood Forecasting Model Based on Artificial Neural Networks," Journal of The Korean Society of Agricultural Engineers, Vol.63, No.1(2021), 11~25.
Kim, K. J. and H. C. Ahn, "Optimization of Support Vector Machines for Financial Forecasting," Journal of Intelligence and Information Systems, Vol.17, No.4(2011), 241~254.
Kim, K. Y., G. R. Lee, and S.W. Lee, "A Comparative Analysis of Artificial Intelligence System and Ohlson model for IPO firm's Stock Price Evaluation," Journal of Digital Convergence, Vol.11, No.5(2013), 145~158.
Kim, S. J. and H. C. Ahn, "Application of Random Forests to Corporate Credit Rating Prediction," The Journal of Business and Economics, Vol.32, No.1(2016), 187~211
Kim, T. H., "A Study on the present situations and some proposal for improvements of IPO Regulations and Systems," Journal of Business Administration & Law, Vol.26, No.4(2016), 201~238.
Kim, Y. S. and S. W. Joh, "Text Analysis for IPO firms in Korea: Analysis of Korean Texts in Registration Statements via Machine Learning," Korean Journal of Financial Studies, Vol.48, No.2(2019), 215~235.
Kolchyna, O., T. T. Souza., P. Treleaven., and T. Aste, "Twitter sentiment analysis: Lexicon method, machine learning method and their combination," arXiv preprint arXiv:1507.009 55, (2015).
Lee, H. S., S. H. Jeong, and K. J. Oh, "A study on the prediction of Korean NPL market return," Journal of Intelligence and Information Systems, Vol.25, No.2(2019), 123~139.
Lee, S. W. and J. H. Kim, "A Study on the Extraction of Psychological Distance Embedded in Company's SNS Messages Using Machine Learning," Information Systems Review, Vol.21, No.1(2019), 23~38.

원문보기 상세보기
Li, F, "The Information Content of Forward-Looking Statements in Corporate Filings-A Naive Bayesian Machine Learning Approach," Journal of Accounting Research, Vol.48, No. 5(2010), 1049~1102.

상세보기
Loughran, T. and B. McDonald, "When Is a Liability Not a Liability? Textual Analysis, Dictionaries, and 10-Ks," The Journal of Finance, Vol.66, No.1(2011), 35~65.

상세보기
Loughran, T. and B. McDonald, "IPO first-day returns, offer price revisions, volatility, and form S-1 language," Journal of Financial Economics, Vol.109, No.2(2013), 307~326.

상세보기
Luque, C., D. Quintana., and P. Isasi, "Predicting IPO Underpricing with Genetic Algorithms," International Journal of Artificial Intelligence, Vol.8, No.12(2012), 133~146.
Ly, T. H. and K. Nguyen, "Do Words Matter: Predicting IPO Performance from Prospectus Sentiment," 2020 IEEE 14th International Conference on Semantic Computing, (2020), 307~310.
Mai, F., S. Tian., C. Lee., and L. Ma, "Deep learning models for bankruptcy prediction using textual disclosures," European Journal of Operational Research, Vol.274, No.2(2019), 743~758.

상세보기
Manurung, J., H. Mawengkang., and E. Zamzami, "Optimizing Support Vector Machine Parameters with Genetic Algorithm for Credit Risk Assessment," Journal of Physics: Conference Series, Vol.930, No. 1(2017).
Martens, D. and F. Provost, "Explaining Data-Driven Document Classifications," MIS Quarterly, Vol.38, No.1(2014), 73-100.

상세보기
Miller, E. M, "Risk, uncertainty, and divergence of opinion," Journal of Finance, Vol.32, No.4 (1977), 1151~1168.

상세보기
Min, J. H, "IPO Stock Performance of Institutional and Individual Investors," Journal of CEO and Management Studies, Vol.20, No.3(2017), 75~98.
Moraes, R., J. F. Valiati, and W. P. G. Neto, "Document-level sentiment classification: An empirical comparison between SVM and ANN," Expert Systems with Applications, Vol.40, No.2(2013), 621-633.

상세보기
Muditomo, A. and A. S. Broto, "IPO Performance Prediction During Covid-19 Pandemic in Indonesia Using Decision Tree Algorithm," Journal of Finance and Banking, Vol.25, No.1(2021), 132~143.
Nguyen, H., A. Veluchamy., M. Diop., and R. Iqbal, "Comparative Study of Sentiment Analysis with Product Reviews Using Machine Learning and Lexicon-Based Approaches," SMU Data Science Review, Vol.1, No.4(2018), 7.
Park J. S. and B. H. Han, "Analyzing the IPO Market of 2020 and Forecasting the Market of 2021," KOSDAQ Venture, Eugene Research Center, (2021), 16-150.
Park, J. W., G. C. Jung, and J. E. Cho, "Institutional Investor Trading and IPOs Performance," Korean Journal of Financial Studies, Vol.45, No.1(2016), 171~192.
Park K. J. and J. Q. Jeon, "The Effect of IPO Syndicates on Underwriting Services: Focusing on Multiple Lead Underwriters and Co-Managers," Korean Journal of Financial Studies, Vol.44, No.1(2015), 189~219.
Park, S. G., H. J. Lee, H. J. Sim, J. Y. Lee, and J. E. Oh, "Construction of Sound Quality Index for the Vehicle HVAC System Using Regression Model and Neural Network Model," Korean Society for Noise and Vibration Engineering, (2006), 1308~1313.
Perera, W. and N. Kulendran, "Short-run underpricing and its determinants: Evidence from Australian IPOS," Corporate Ownership and Control, Vol.13, No.3(2016), 502~517.

상세보기
Prastyo, P.H., I. Ardiyanto., and R. Hidayat, "Indonesian Sentiment Analysis: An Experimental Study of Four Kernel Functions on SVM Algorithm with TF-IDF," 2020 International Conference on Data Analytics for Business and Industry: Way Towards a Sustainable Economy (ICDABI), (2020), 1~6.
Quintana, D., F. Chavez., R. M. Luque Baena., and F. Luna, "Fuzzy techniques for IPO underpricing prediction," Journal of Intelligent & Fuzzy Systems, Vol.35, No.1(2018), 367~381.

상세보기
Reber, B., B. Berry., and S. Toms, "Predicting mispricing of initial public offerings," Intelligent Systems in Accounting, Finance, and Management, Vol.13, No.1(2005), 41~59.

상세보기
Rock, K, "Why new issues are underpriced," Journal of Financial Economics, Vol.15, No.1-2 (1986), 187~212.
Ruud, J. S, "Underwriter price support and the IPO underpricing puzzle," Journal of Financial Economics, Vol.34, No.2(1993), 135~151.

상세보기
Seo, K. K., "Sales Prediction of Electronic Appliances using a Convergence Model based on Artificial Neural Network and Genetic Algorithm," Journal of Digital Convergence, Vol.13, No.9(2015), 177~182.
Seo, S. H. and J. T. Kim "Research Trends in Deep Learning-Based Sentiment Analysis," Journal of Korea Multimedia Society, Vol.20, No.3(2016), 8~22.
Shin, S. H., H. J. Lee, and J. J. Ahn, "A study on initial price change prediction of IPO shares using non-financial information," Journal of the Korean Data and Information Science Society, Vol.29, No.2(2018), 589~616.
Sun, A., E. P. Lim, and Y. Liu, "On strategies for imbalanced text classification using SVM: A comparative study," Decision Support Systems, Vol.48, No.1(2009), 191-201.

상세보기
Tao, J., A. V. Deokar., and A. Deshmukh, "Analysing forward-looking statements in initial public offering prospectuses: a text analytics approach," Journal of Business Analytics, Vol.1, No.1(2018), 54~70.

상세보기
Tetlock, P. C., M. Saar-Tsechansky., and S. Macskassy, "More than words: Quantifying language to measure firms' fundamentals," The Journal of Finance, Vol.63, No.3(2008), 1437~1467.

상세보기
Yan, Y., X. Xiong., J. G. Meng., & G. Zou, "Uncertainty and IPO initial returns: Evidence from the Tone Analysis of China's IPO Prospectuses," Pacific-Basin Finance Journal, Vol.57, (2019), 101075.

상세보기

저자의 다른 논문 :

표제어: PCR

동의어: Packet Collision Rate

용어 설명 출처 목록 (6)

용어 설명: PCR은 세균 특이성이 있는 primer를 이용하여 적은 수의 세균이 있을지라도 쉽게 검출할 수 있는 유용한 방법이며, 이를 이용하여 구강 내 치면세균막이나 타액에서 직접 세균을 검출할 수 있게 되었다[8].

내보내기 구분	파일저장 인쇄 메일전송
구성항목	기본정보 상세정보 관리번호, 논문명, 저널/프로시딩명, 저자 , 발행년, 권, 호, 시작페이지, 끝페이지, 발행기관 관리번호, 논문명, 대등논문명, 저자 , 저널/프로시딩명, 발행기관, 발행년, 발행언어, 권, 호, 시작페이지, 끝페이지, ISBN, ISSN, 주제분야, 키워드, 초록(한글), 초록(영문), 저자(소속기관)
저장형식	Text(ASCII format) Excel format RefWorks Direct Export RIS format (for Reference Manager, ProCite, EndNote), Scholar's Aids, Mendeley
메일정보	받는사람 (필수) @ 보내는사람 (선택) @ 제목 내용 KISTI 검색결과 이메일 서비스
안내	총 건의 자료가 검색되었습니다. 다운받으실 자료의 인덱스를 입력하세요. (1-10,000) 검색결과의 순서대로 최대 10,000건 까지 다운로드가 가능합니다. 데이타가 많을 경우 속도가 느려질 수 있습니다.(최대 2~3분 소요) 다운로드 파일은 UTF-8 형태로 저장됩니다. 파일의 내용이 제대로 보이지 않을실 때는 웹브라우저 상단의 보기 -> 인코딩 -> 자동선택 여부를 확인하십시오. ~ Text(ASCII format) Excel format

연합인증