Copper and Aluminium as Economically Imperfect Substitutes, Production and Price Development

Kód výsledku v IS VaVaI

<a href="https://www.isvavai.cz/riv?ss=detail&h=RIV%2F00216305%3A26510%2F22%3APU145328" target="_blank" >RIV/00216305:26510/22:PU145328 - isvavai.cz</a>

Nalezeny alternativní kódy

RIV/75081431:_____/22:00002371

Výsledek na webu

<a href="https://actamont.tuke.sk/pdf/2022/n2/14bartos.pdf" target="_blank" >https://actamont.tuke.sk/pdf/2022/n2/14bartos.pdf</a>

DOI - Digital Object Identifier

<a href="http://dx.doi.org/10.46544/AMS.v27i2.14" target="_blank" >10.46544/AMS.v27i2.14</a>

Jazyk výsledku

angličtina

Název v původním jazyce

Copper and Aluminium as Economically Imperfect Substitutes, Production and Price Development

Popis výsledku v původním jazyce

Copper and aluminium prices have long been influenced mainly by non-renewable resources and the industry's widespread use of copper and aluminium for their desired properties. Metal commodities are irreplaceable for the industry of developed countries, and their shortage in the covid times also increases the price and consequently the price of products made from them. As copper ore stocks continue to decline, suitable substitutes should be sought. The paper discusses the potential of copper substitution by aluminium and subsequently the development of prices and production of copper and aluminium, including a prediction about the future development. Research data were obtained from Market. business insider (2021) and Investing.com (2022) converted to time series. The price is shown in US dollars per tonne and the production value in millions of tonnes. Development data were processed using artificial intelligence and recurrent neural networks, including the Long Short Term Memory layer. Neural networks, as such, have great potential to predict these types of time series. The annual copper and aluminium production data were processed using a regression function. Neural networks could not be used due to the smaller data range. The results show that the 1NN30L neural network with an LSTM layer and considered a 30-day delay is the most suitable network for forecasting future copper prices, and the 3NN30L neural network with an LSTM layer and considered a 30-day delay is the most suitable network for forecasting future aluminium prices. The forecast has confirmed that the price of copper will fall at the end of 2021, and the trend will be constant in the next planned period. Aluminium will also fall sharply at the end of 2021; at the beginning of 2022, the price level is predicted to rise to that of 30 October 2021, and thereinafter the trend will be almost constant. Research has confirmed that copper and aluminium may be imperfect substitutes in some respect, but they c

Název v anglickém jazyce

Copper and Aluminium as Economically Imperfect Substitutes, Production and Price Development

Popis výsledku anglicky

Copper and aluminium prices have long been influenced mainly by non-renewable resources and the industry's widespread use of copper and aluminium for their desired properties. Metal commodities are irreplaceable for the industry of developed countries, and their shortage in the covid times also increases the price and consequently the price of products made from them. As copper ore stocks continue to decline, suitable substitutes should be sought. The paper discusses the potential of copper substitution by aluminium and subsequently the development of prices and production of copper and aluminium, including a prediction about the future development. Research data were obtained from Market. business insider (2021) and Investing.com (2022) converted to time series. The price is shown in US dollars per tonne and the production value in millions of tonnes. Development data were processed using artificial intelligence and recurrent neural networks, including the Long Short Term Memory layer. Neural networks, as such, have great potential to predict these types of time series. The annual copper and aluminium production data were processed using a regression function. Neural networks could not be used due to the smaller data range. The results show that the 1NN30L neural network with an LSTM layer and considered a 30-day delay is the most suitable network for forecasting future copper prices, and the 3NN30L neural network with an LSTM layer and considered a 30-day delay is the most suitable network for forecasting future aluminium prices. The forecast has confirmed that the price of copper will fall at the end of 2021, and the trend will be constant in the next planned period. Aluminium will also fall sharply at the end of 2021; at the beginning of 2022, the price level is predicted to rise to that of 30 October 2021, and thereinafter the trend will be almost constant. Research has confirmed that copper and aluminium may be imperfect substitutes in some respect, but they c

Druh

J_imp - Článek v periodiku v databázi Web of Science

CEP obor

—

OECD FORD obor

20701 - Environmental and geological engineering, geotechnics

Projekt

—

Návaznosti

S - Specificky vyzkum na vysokych skolach

Rok uplatnění

2022

Kód důvěrnosti údajů

S - Úplné a pravdivé údaje o projektu nepodléhají ochraně podle zvláštních právních předpisů

Název periodika

Acta Montanistica Slovaca

ISSN

1335-1788

e-ISSN

1339-3103

Svazek periodika

27

Číslo periodika v rámci svazku

2

Stát vydavatele periodika

SK - Slovenská republika

Počet stran výsledku

17

Strana od-do

462-478

Kód UT WoS článku

000834987800013

EID výsledku v databázi Scopus

2-s2.0-85135301476