Predicting Stock Price Movements with Text Data using Labeling based on Financial Theory

Fredrik Ahnve; Kasper Fantenberg; Gustav Svensson; Daniel Hardt

doi:10.1109/BigData50022.2020.9378054

Predicting Stock Price Movements with Text Data using Labeling based on Financial Theory

Fredrik Ahnve, Kasper Fantenberg, Gustav Svensson, Daniel Hardt^*

^*Corresponding author for this work

Department of Management, Society and Communication

Research output: Chapter in Book/Report/Conference proceeding › Article in proceedings › Research › peer-review

Abstract

We apply Natural Language Processing and supervised Machine Learning to predict stock price movements, based on approximately 30,000 ad-hoc disclosures issued by publicly traded companies on Nasdaq OMX Stockholm. Three different labeling methods, based on financial theory, are defined and assessed. The best results, using Logistic Regression and TF-IDF with character-grams, achieve an increase of 6,3 percentage points above a majority class baseline. These results show that corporate ad-hoc disclosures, which are regulated to represent novel and value-relevant information, are particularly well-suited for this task. Furthermore, the most sophisticated labeling technique used, Jensen’s Alpha in the context of the Capital Asset Pricing Model, helps the model achieve its highest accuracy. The results therefore show that financial theory can help isolate the effect of an informational event on stock prices, improving the supervised Machine Learning approach. Finally, an algorithmic trading strategy is simulated with the best model, yielding positive abnormal returns.

Original language	English
Title of host publication	Proceedings - 2020 IEEE International Conference on Big Data. Big Data 2020
Editors	Xintao Wu, Chris Jermaine, Li Xiong, Xiaohua Hu, Olivera Kotevska, Siyuan Lu, Weija Xu, Srinivas Aluru, Chengxiang Zhai, Eyhab Al-Masri, Zhiyuan Chen, Jeff Saltz
Number of pages	8
Place of Publication	Los Alamitos, CA
Publisher	IEEE
Publication date	2020
Pages	4365-4372
Article number	9378054
ISBN (Print)	9781728162522
ISBN (Electronic)	9781728162515
DOIs	https://doi.org/10.1109/BigData50022.2020.9378054
Publication status	Published - 2020
Event	Eighth IEEE International Conference on Big Data. IEEE BigData 2020 - Virtual Event Duration: 10 Dec 2020 → 13 Dec 2020 Conference number: 8 https://bigdataieee.org/BigData2020/

Conference

Conference	Eighth IEEE International Conference on Big Data. IEEE BigData 2020
Number	8
Location	Virtual Event
Period	10/12/2020 → 13/12/2020
Internet address	https://bigdataieee.org/BigData2020/

Keywords

Algorithmic trading
Machine learning
Stock price prediction
Ad-hoc disclosures
Natural language processing
Text mining
Finance

Access to Document

10.1109/BigData50022.2020.9378054Licence: Unspecified

Handle.net

Persistent link

Cite this

Ahnve, F., Fantenberg, K., Svensson, G., & Hardt, D. (2020). Predicting Stock Price Movements with Text Data using Labeling based on Financial Theory. In X. Wu, C. Jermaine, L. Xiong, X. Hu, O. Kotevska, S. Lu, W. Xu, S. Aluru, C. Zhai, E. Al-Masri, Z. Chen, & J. Saltz (Eds.), Proceedings - 2020 IEEE International Conference on Big Data. Big Data 2020 (pp. 4365-4372). Article 9378054 IEEE. https://doi.org/10.1109/BigData50022.2020.9378054

Ahnve, Fredrik ; Fantenberg, Kasper ; Svensson, Gustav et al. / Predicting Stock Price Movements with Text Data using Labeling based on Financial Theory. Proceedings - 2020 IEEE International Conference on Big Data. Big Data 2020. editor / Xintao Wu ; Chris Jermaine ; Li Xiong ; Xiaohua Hu ; Olivera Kotevska ; Siyuan Lu ; Weija Xu ; Srinivas Aluru ; Chengxiang Zhai ; Eyhab Al-Masri ; Zhiyuan Chen ; Jeff Saltz. Los Alamitos, CA : IEEE, 2020. pp. 4365-4372

@inproceedings{9e7ded3cfa4e4d419203ecf20bfae822,

title = "Predicting Stock Price Movements with Text Data using Labeling based on Financial Theory",

abstract = "We apply Natural Language Processing and supervised Machine Learning to predict stock price movements, based on approximately 30,000 ad-hoc disclosures issued by publicly traded companies on Nasdaq OMX Stockholm. Three different labeling methods, based on financial theory, are defined and assessed. The best results, using Logistic Regression and TF-IDF with character-grams, achieve an increase of 6,3 percentage points above a majority class baseline. These results show that corporate ad-hoc disclosures, which are regulated to represent novel and value-relevant information, are particularly well-suited for this task. Furthermore, the most sophisticated labeling technique used, Jensen{\textquoteright}s Alpha in the context of the Capital Asset Pricing Model, helps the model achieve its highest accuracy. The results therefore show that financial theory can help isolate the effect of an informational event on stock prices, improving the supervised Machine Learning approach. Finally, an algorithmic trading strategy is simulated with the best model, yielding positive abnormal returns.",

keywords = "Algorithmic trading, Machine learning, Stock price prediction, Ad-hoc disclosures, Natural language processing, Text mining, Finance, Algorithmic trading, Machine learning, Stock price prediction, Ad-hoc disclosures, Natural language processing, Text mining, Finance",

author = "Fredrik Ahnve and Kasper Fantenberg and Gustav Svensson and Daniel Hardt",

year = "2020",

doi = "10.1109/BigData50022.2020.9378054",

language = "English",

isbn = "9781728162522",

pages = "4365--4372",

editor = "Xintao Wu and Chris Jermaine and Li Xiong and Xiaohua Hu and Olivera Kotevska and Siyuan Lu and Weija Xu and Srinivas Aluru and Chengxiang Zhai and Eyhab Al-Masri and Zhiyuan Chen and Jeff Saltz",

booktitle = "Proceedings - 2020 IEEE International Conference on Big Data. Big Data 2020",

publisher = "IEEE",

address = "United States",

note = "Eighth IEEE International Conference on Big Data. IEEE BigData 2020 ; Conference date: 10-12-2020 Through 13-12-2020",

url = "https://bigdataieee.org/BigData2020/",

}

Ahnve, F, Fantenberg, K, Svensson, G & Hardt, D 2020, Predicting Stock Price Movements with Text Data using Labeling based on Financial Theory. in X Wu, C Jermaine, L Xiong, X Hu, O Kotevska, S Lu, W Xu, S Aluru, C Zhai, E Al-Masri, Z Chen & J Saltz (eds), Proceedings - 2020 IEEE International Conference on Big Data. Big Data 2020., 9378054, IEEE, Los Alamitos, CA, pp. 4365-4372, Eighth IEEE International Conference on Big Data. IEEE BigData 2020, 10/12/2020. https://doi.org/10.1109/BigData50022.2020.9378054

Predicting Stock Price Movements with Text Data using Labeling based on Financial Theory. / Ahnve, Fredrik; Fantenberg, Kasper; Svensson, Gustav et al.
Proceedings - 2020 IEEE International Conference on Big Data. Big Data 2020. ed. / Xintao Wu; Chris Jermaine; Li Xiong; Xiaohua Hu; Olivera Kotevska; Siyuan Lu; Weija Xu; Srinivas Aluru; Chengxiang Zhai; Eyhab Al-Masri; Zhiyuan Chen; Jeff Saltz. Los Alamitos, CA: IEEE, 2020. p. 4365-4372 9378054.

Research output: Chapter in Book/Report/Conference proceeding › Article in proceedings › Research › peer-review

TY - GEN

T1 - Predicting Stock Price Movements with Text Data using Labeling based on Financial Theory

AU - Ahnve, Fredrik

AU - Fantenberg, Kasper

AU - Svensson, Gustav

AU - Hardt, Daniel

N1 - Conference code: 8

PY - 2020

Y1 - 2020

N2 - We apply Natural Language Processing and supervised Machine Learning to predict stock price movements, based on approximately 30,000 ad-hoc disclosures issued by publicly traded companies on Nasdaq OMX Stockholm. Three different labeling methods, based on financial theory, are defined and assessed. The best results, using Logistic Regression and TF-IDF with character-grams, achieve an increase of 6,3 percentage points above a majority class baseline. These results show that corporate ad-hoc disclosures, which are regulated to represent novel and value-relevant information, are particularly well-suited for this task. Furthermore, the most sophisticated labeling technique used, Jensen’s Alpha in the context of the Capital Asset Pricing Model, helps the model achieve its highest accuracy. The results therefore show that financial theory can help isolate the effect of an informational event on stock prices, improving the supervised Machine Learning approach. Finally, an algorithmic trading strategy is simulated with the best model, yielding positive abnormal returns.

AB - We apply Natural Language Processing and supervised Machine Learning to predict stock price movements, based on approximately 30,000 ad-hoc disclosures issued by publicly traded companies on Nasdaq OMX Stockholm. Three different labeling methods, based on financial theory, are defined and assessed. The best results, using Logistic Regression and TF-IDF with character-grams, achieve an increase of 6,3 percentage points above a majority class baseline. These results show that corporate ad-hoc disclosures, which are regulated to represent novel and value-relevant information, are particularly well-suited for this task. Furthermore, the most sophisticated labeling technique used, Jensen’s Alpha in the context of the Capital Asset Pricing Model, helps the model achieve its highest accuracy. The results therefore show that financial theory can help isolate the effect of an informational event on stock prices, improving the supervised Machine Learning approach. Finally, an algorithmic trading strategy is simulated with the best model, yielding positive abnormal returns.

KW - Algorithmic trading

KW - Machine learning

KW - Stock price prediction

KW - Ad-hoc disclosures

KW - Natural language processing

KW - Text mining

KW - Finance

KW - Algorithmic trading

KW - Machine learning

KW - Stock price prediction

KW - Ad-hoc disclosures

KW - Natural language processing

KW - Text mining

KW - Finance

U2 - 10.1109/BigData50022.2020.9378054

DO - 10.1109/BigData50022.2020.9378054

M3 - Article in proceedings

SN - 9781728162522

SP - 4365

EP - 4372

BT - Proceedings - 2020 IEEE International Conference on Big Data. Big Data 2020

A2 - Wu, Xintao

A2 - Jermaine, Chris

A2 - Xiong, Li

A2 - Hu, Xiaohua

A2 - Kotevska, Olivera

A2 - Lu, Siyuan

A2 - Xu, Weija

A2 - Aluru, Srinivas

A2 - Zhai, Chengxiang

A2 - Al-Masri, Eyhab

A2 - Chen, Zhiyuan

A2 - Saltz, Jeff

PB - IEEE

CY - Los Alamitos, CA

T2 - Eighth IEEE International Conference on Big Data. IEEE BigData 2020

Y2 - 10 December 2020 through 13 December 2020

ER -

Ahnve F, Fantenberg K, Svensson G, Hardt D. Predicting Stock Price Movements with Text Data using Labeling based on Financial Theory. In Wu X, Jermaine C, Xiong L, Hu X, Kotevska O, Lu S, Xu W, Aluru S, Zhai C, Al-Masri E, Chen Z, Saltz J, editors, Proceedings - 2020 IEEE International Conference on Big Data. Big Data 2020. Los Alamitos, CA: IEEE. 2020. p. 4365-4372. 9378054 doi: 10.1109/BigData50022.2020.9378054