Ivlev A.S. Non-ferrous metal prices forecast on the london metal exchange using a neural network

Details

Parent Category: Geo-Technical Mechanics, 2022

Category: Geo-Technical Mechanics, 2022, № 160

Geoteh. meh. 2022, 160, 29-32

https://doi.org/10.15407/geotm2022.160.029

 

NON-FERROUS METAL PRICES FORECAST ON THE LONDON METAL EXCHANGE USING A NEURAL NETWORK

¹Ivlev A.S.

¹Institute of Geotechnical Mechanics named by N. Poljakov of National Academy of Sciences of Ukraine

UDC 338.5.018.2:004.9

Language: English

Abstract. Any enterprises specializing in the mining industry deal directly with a variety of minerals that are located deep under the thickness of the earth's crust. Among these minerals, ores of non-ferrous metals such as copper, aluminum, zinc, iron, titanium, etc. are also often found. These ores undergo primary processing after the mining stage. Enrichment, processing, smelting of the non-ferrous metals themselves and their various alloys, as well as the manufacture of various products or raw materials necessary for use in other branches of industry, can follow. Certain non-ferrous metals play an important role in the further chain of the production process. As a result, finished products of metal processing or raw materials can become an object of import-export on the international market. The most interesting market is the non-ferrous metals themselves due to the complexity of their extraction and value for the industry as a whole. In order to make the most profitable purchase of non-ferrous metal raw materials on the international exchange, it is necessary to know the exact price of this metal at a certain point in the future, when the cost of the metal at the time of purchase will be as cheap as possible. But, since the future price is practically unknown in advance, there is a need to forecast the price of a certain non-ferrous metal, at least in the short term. In this article, a software device based on the work of a neural network is developed, which predicts the price of some non-ferrous metals for a short period of time in the future based on already existing data on the value of these metals. As an example of the work of the neural network, the article uses a temporary series of prices for iron, aluminum and zinc on the London Metal Exchange, which has accurate data at its disposal with an interval of 1-3 days. The proposed neural network device performs cost forecasting based on a nonlinear autoregression algorithm with pre-set neural network parameters. The developed forecasting method makes it possible to determine the price of non-ferrous metal with a certain degree of accuracy for its further purchase at a price favorable to the buyer.
Keywords: forecast, neural network, training, accuracy, time series.

REFERENCES:

1. London Metal Exchange (2021), “LME Alluminium”, available at: https://www.lme.com/Metals/Non-ferrous/Aluminium#tabIndex=2 (Accessed 27.06.2021).

2. London Metal Exchange (2021), “LME Copper”, available at: https://www.lme.com/Metals/Non-ferrous/Copper#tabIndex=2 (Accessed 27.06.2021).

3. London Metal Exchange (2021), “LME Zinc”, available at: https://www.lme.com/Metals/Non-ferrous/Zinc#tabIndex=2 (Accessed 27.06.2021).

4. Lomakin, N.I., Maximova, O. N., Ekova, V.A., Gavrilova, O.A.  and Vagina, V.E. (2016), “Neural networks for predicting the dollar rate using astrological cyclical indices Gushon and Ganu”, International Journal of Applied and Fundamental Research, no. 6 (1), pp. 133-136.

5. Parisi, A., Parisi, F. and Diaz, D. (2008), “Forecasting changes in the price of gold: convolutional and regression models of neural networks”, Journal of international financial management, no. 18, pp. 477–487. https://doi.org/10.1016/j.mulfin.2007.12.002

6. Fedorova, E.A. and Linkova, M.A. (2013), “Forecasting the exchange rate using neural networks”, Monetary policy, no. 11 (149), pp. 27–31.

7. Ivlev, A.S. (2019), “The use of a neural network in the task of predicting the rate of precious metals”, Matematychne ta programne zabezpechennya intelektualnych system: Tezy dopovidey XVІІ Mizhnarodnoy naukovo-praktychnoy konferenstii MPZIC-2019 [Mathematical and software security of intellectual systems: Abstracts of the XVII International Scientific and Practical Conference MPZIS-2019], Dnipro, 20-22 leaf fall 2019 r. / Edited by O.M. Kiselyovoy, DNU, Dnipro, Ukraine, pp. 112-113.

8. Belozerov, V.E. and Ivlev, A.S. (2019), “Neural network approach for modeling prices in the precious metals market”, Questions of applied mathematics and mathematical modeling: coll. Science, О.М. Kiselyova (ed.), vol. 19, pp. 17–23. https://doi.org/10.15421/321902

About author:

Ivlev Andrii Serhiiovich, Master of Science (M.Sc.), Junior Researcher of Geodynamic Systems and Vibration Technologies Department, Institute of Geotechnical Mechanics named by N. Poljakov of National Academy of Sciences of Ukraine (IGTM NAS of Ukraine), Dnipro, Ukraine, This email address is being protected from spambots. You need JavaScript enabled to view it.

• < Prev
• Next >