Журнал «Современная Наука»

Russian (CIS)English (United Kingdom)
 MOSCOW +7(495)-142-86-81

TIME SERIES FORECASTING MODEL BASED ON THE EXAMPLE OF CURRENCY QUOTATIONS

Khabarov Mark Eduardovich  (Bauman Moscow State Technical University, Moscow)

Lovtsov Vladimir Alekseevich  (Bauman Moscow State Technical University, Moscow)

Tiazhelnikova Ksenia Sergeevna  (Head of projects in TBank JSC, Moscow)

Gurenko Vladimir Viktorovich  (Associate Professor, Bauman Moscow State Technical University, Moscow)

The article describes the actual task of developing a model of time series forecasting on the example of quotes of currency pairs. The peculiarities of working with time series are defined and the mathematical model of the subject area of the developed model is formulated. The process of dividing a time series into components is analyzed. The main approaches to the realization of time series forecasting model are studied, namely: statistical algorithms and machine learning. Necessary and sufficient conditions of stationarity of time series for realization of the model of statistical approach are considered. The initial data set for the development of time series forecasting model on the example of currency quotes based on the dataset “Currency rates: archival and current data on the value of foreign currencies against the ruble” was formed. A set of lag and aggregated informative features for the realization of models based on machine learning was formed. Practical experiments were conducted to analyze the behavior of the models in different situations of financial market stability. The results of each algorithm within the conducted experiments are presented, including MSE, MAE, MAPE and R2 metrics. Based on the results obtained, the optimal algorithm was selected to realize the task at hand. Also, the most informative features of this model for forecasting time series of currency quotes were determined.

Keywords:time series, machine learning, time series forecasting, forecasting approaches, quotes, currency pairs.
 

Read the full article …

Citation link:
Khabarov M. E., Lovtsov V. A., Tiazhelnikova K. S., Gurenko V. V. TIME SERIES FORECASTING MODEL BASED ON THE EXAMPLE OF CURRENCY QUOTATIONS // Современная наука: актуальные проблемы теории и практики. Серия: Естественные и Технические Науки. -2025. -№04. -С. 164-169 DOI 10.37882/2223-2966.2025.04.45
LEGAL INFORMATION:
Reproduction of materials is permitted only for non-commercial purposes with reference to the original publication. Protected by the laws of the Russian Federation. Any violations of the law are prosecuted.
© ООО "Научные технологии"

Terms Publications

Issues of the journals (Archive)

ECONOMICS AND LAW
HUMANITIES
NATURAL AND TECHNICAL SCIENCES
COGNITION

Серия - Natural and Technical Sciences