Improving the accuracy of prediction of foreign exchange rates on the internet market by means of neural networks

User Rating:  / 0
PoorBest 

Authors:

Ye.Yu. Churikanova, Cand. Sci. (Econ.), Associate Professor, State Higher Educational Institution “National Mining University”, Senior Lecturer of the Department of Economic Cybernetics and Information Technologies, Dnipropetrovsk, Ukraine

K.Ye. Tukhmenova, State Higher Educational Institution “National Mining University”, Assistant Lecturer of the Management of Production Sphere Department, Dnipropetrovsk, Ukraine

M.A. Kharchenko,State Higher Educational Institution “National Mining University”, postgraduate student, Dnipropetrovsk, Ukraine

Abstract:

Purpose. Development and construction of economic-mathematical forecasting model of exchange rates by means of neural networks for definition of exchange rates behavior on the Internet market taking into account factors of technical and fundamental analyses.

Methodology. For solution of this goal methods of the comparative analysis, systematic comprehensive approach and methods of economic-mathematical modeling by means of neural networks were used.

Findings. Research of major factors by tools available to do a forecast concerning fluctuations of exchange rates in the future was carried out. We have defined that the exchange rates behavior is influenced by rather large number of factors relating to methods of fundamental and technical analyses. It was established that any factors cannot guarantee 100% reliability in a forecast. Only application of a comprehensive approach insures high degree of forecast accuracy. Economic-mathematical forecasting model of exchange rates with the neural networks for the purpose of implementation of a comprehensive approach was offered and constructed. It has allowed complex indicators of fundamental and technical analyses. On the basis of technical and fundamental analyses indicators of exchange rates behavior on the Internet market the group of factors which have numerical measurement was created. Justification of mathematical apparatus of neural networks as optimum for construction economic-mathematical model for the performance of forecasts and taking into account the purpose put in this work was executed. On the basis of the created group of factors construction of economy was executed - mathematical model of funds of neural networks, for which entrance data on the selected factors move and on exit look-ahead value of exchange rates turns out.

Originality. Construction of economic-mathematical forecasting model of exchange rates for Internet market was executed by neural networks which unlike the existing considering groups of ten factors, such as: the index of the relative size of the prices, the index of simplification of the market, the index of Dow Jones, the “Standard and powers” index, the New York stock exchange index, indices of the American stock exchange, the RSI indicator, the average index of the directed movement ADX, the index of off-exchange turn, the stochastic indicator.

Practical value. The practical significance of these results is the possibility of increasing the accuracy of the forecast of exchange rates on the Internet market by the use of advanced mathematical apparatus, which was based on the expanded set of factors affecting the fluctuations of exchange rates.

References:

1. Urintsov, A.I. (2003), “Regarding usage of the EPSS system as the instrument for on-line adaptation of the Decision Maker”, Pribory i Systemy Upravleniya, no. 11, pp. 63–66.

Уринцов A.И. Regarding usage of the EPSS system as the instrument for on-line adaptation of the Decision Maker/ A.И. Уринтсов. // Приборы и системы управления. – № 11. –  2003. – С. 63–66.

2. Alpesh B. Patel (2003), Trading Online: A Step-by-Step Guide to CyberProfits, Williams, UK.

Пейтел Э. Internet-трейдинг. Полное руководство / Э. Пейтл, П. Пейтл;  пер. с англ. – М.: Издательский дом „Вильямс“, 2003. – 320 с.

3. Taran, V.A. (2003), Playing the stock market is easy!, NP “Forex club”, Forex Club, Moscow, Russia.

Таран B.A. Играть на бирже просто?! / Таран B.A.  // НП Форекс клуб. – М.: Форекс Клуб, 2003. –240с.

4. Robert W. Colby and Thomas A. Meyers (1998), The Encyclopedia of Technical Market Indicators, Dow Jones-Irwin.

Роберт В.Колби. Р. Энциклопедия технических индикаторов рынка / В. Роберт, А. Мейерс – М.: Издательский дом Альпина, 1998. 581 с.

5. T. Demark (1997), Tekhnicheskiy analiz, novaya nauka [Technical Analysis, New Science], Diagramma, Moscow, Russia.

Демарк Т. Технический анализ, новая наука / Демарк Т. – М.: Диаграмма, 1997. – 178 с.

6. Yezhov, A.A. and Shumsky, S.A. (1998), Neirokompyuting i yego primeneniye v ekonomike i biznese [Neurocomputing and Its Application in Economics and Business], MIFI, Moscow, Russia.

Ежов А.А. Нейрокомпьютинг и его применение в экономике и бизнесе / А.А. Ежов, С.А. Шумски – М.: МИФИ, 1998.  – 216с. 

7. Minayev, Yu.N., Filimonova, O.Yu. and Benameur, L. (2003), Metody i  algoritmy resheniya zadach identifikatsyi i prognozirovaniya v usloviyakh neopredelennosti v neirosetevom logicheskom bazise [Methods and Algorithms for Solving Identification and Forecasting Tasks Under Uncertainty in the Neural Network Logical Basis], Hotline Telecom, Moscow, Russia.

Минаев Ю.Н. Методы и алгоритмы решения задач идентификации и прогнозирования в условиях неопределенности в нейросетевом логическом базисе / Минаев Ю.Н., Филимонова О.Ю., Бенамеур Л. – М.: Изд-во. Горячая линия-Телеком, 2003. – 205 с.

8. Callan, R. (1999), The Essence of Neural Networks, Prentice Hall Europe.

Каллан Р. Основные концепции нейронных сетей / Каллан Р. – М.: Изд-во. Вильямс, 2002. – 287 с.

9. Rutkovska, D. (2004), Neironnye seti, geneticheskiye algoritmy i nechetkiye sistemy [Neural networks, genetic algorithms and fuzzy systems], Hotline Telecom, Moscow, Russia.

Рутковская Д. Нейронные сети, генетические алгоритмы и нечёткие системы / Рутковская Д. – М.: Изд-во. Горячая линия-Телеком, 2004. – 452с.

10. Haykin, S. (2006), Neironnye seti [Neural Networks: a Complete Course] Dialektika, Kiyev, Ukraine.

Хайкин С. Нейронные сети: полный курс / Хайкин С. – К.: Изд-во. Диалектика, 2006. – 1104 с.

Files:
2014_4_churikanova
Date 2014-09-17 Filesize 131.76 KB Download 695

Visitors

3256375
Today
This Month
All days
167
9009
3256375

Guest Book

If you have questions, comments or suggestions, you can write them in our "Guest Book"

Registration data

ISSN (print) 2071-2227,
ISSN (online) 2223-2362.
Journal was registered by Ministry of Justice of Ukraine.
Registration number КВ No.17742-6592PR dated April 27, 2011.

Contacts

D.Yavornytskyi ave.,19, pavilion 3, room 24-а, Dnipro, 49005
Tel.: +38 (056) 746 32 79.
e-mail: This email address is being protected from spambots. You need JavaScript enabled to view it.
You are here: Home Archive by field of science Economy Improving the accuracy of prediction of foreign exchange rates on the internet market by means of neural networks