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

Student
Title
Supervisor
Faculty
Educational Programme
Final Grade
Year of Graduation
Viktor Zhemchuzhnikov
Prediction of MICEX Index: Predictive Power of Machine Learning Methods and Econometric Models
Economics
(Bachelor’s programme)
10
2017
The ability to predict the dynamics of financial instruments is an urgent task for participants in the financial market. In the conditions of a large flow of heterogeneous information, there is a need to use effective methods of processing them for the development of operational management decisions. In particular, the methods of machine learning are becoming increasingly widespread in financial modeling, but there are already a number of econometric models used for such a task. The aim of the work is to compare the ARIMA, ARIMA-GARCH and ARIMA-TGARCH models with neural networks and the method of reference vectors for determining the greatest predictive power in the task of predicting the MICEX index. Information base of economic and mathematical modeling was made by statistical and analytical data on the dynamics of the MICEX index, fundamental and technical indicators of the stock market for 2003 - 2017. The results of the computer experiments were performed on the training and testing sample using the appropriate machine learning libraries in the Python language and the Eviews software package. The estimation of the predictive power of the methods was carried out on a test sample using traditional indicators of mathematical statistics (absolute and relative prediction errors) and the determination coefficient, as well as benchmarks. The test sample shows a higher predictive accuracy of econometric models. Also, ARIMA-GARCH and the support vector method are tested in the proposed simplest trading strategy. ARIMA-GARCH was able to achieve profits greater than the "Buy and Hold" strategy, while the support vector method demonstrated a negative return. Possible areas for further research include the creation of more integrated models that take into account investor sentiment, as well as the development of advanced hybrid models.

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