• A
  • A
  • A
  • ABC
  • ABC
  • ABC
  • А
  • А
  • А
  • А
  • А
Regular version of the site

Model Predicting Stock Market Movements Based on Reported News Developed Using Python

Student: Tarasov Kirill

Supervisor: Boris Pozin

Faculty: HSE Tikhonov Moscow Institute of Electronics and Mathematics (MIEM HSE)

Educational Programme: Computer Systems and Networks (Master)

Final Grade: 10

Year of Graduation: 2017

Key words: Python, Sberbank, Gazprom, Lukoil, Rosneft, machine learning, Twitter, stock movement prediction, tweets classification, information background. In this paper, the prediction of stock market dynamics based on news messages characteristics is being developed and realized. For the study execution, the following key Russian companies were chosen: Sberbank, Lukoil, Gazprom, Rosneft. These companies integrate a significant part of MICEX index. For each of these companies the tweets that contain relevant news have been classified. Classification has been made based on the news influence on the company share price. Two models of classification have been developed: frequency model and dictionary model. For better quality results these two models were combined into one. Further, text features suitable for machine learning algorithm were created based on classified tweets. In order to analyze the relationship between stock market trends and news background three kinds of the stock movement prediction models have been developed: with the use of only technical features (baseline), with both technical and text features, only with text features. For the model based on only text features significant accuracy improvement was noticed compared to baseline model: 22% for Sberbank, 17.4% for Rosneft, 18% for Gazprom and 4% for Lukoil. Then the conclusions on the feasibility to predict stock market dynamics based on news reports have been made. Besides, possible ways of prediction model improvement have been proposed.

Full text (added May 14, 2017)

Student Theses at HSE must be completed in accordance with the University Rules and regulations specified by each educational programme.

Summaries of all theses must be published and made freely available on the HSE website.

The full text of a thesis can be published in open access on the HSE website only if the authoring student (copyright holder) agrees, or, if the thesis was written by a team of students, if all the co-authors (copyright holders) agree. After a thesis is published on the HSE website, it obtains the status of an online publication.

Student theses are objects of copyright and their use is subject to limitations in accordance with the Russian Federation’s law on intellectual property.

In the event that a thesis is quoted or otherwise used, reference to the author’s name and the source of quotation is required.

Search all student theses