Analysis of the Associations Among Cryptocurrencies, Market Indices and Commodities Using Machine Learning
DOI:
https://doi.org/10.20491/isarder.2025.2084Keywords:
Cryptocurrencies, Market Indices and Commodities, Machine Learning, FP-Growth Algorithm, Apriori AlgorithmAbstract
Purpose – Cryptocurrencies, which are becoming increasingly important for both investors and financial systems today, have gained an important place in investors' portfolio strategies due to their high volatility. The aim of the study is to reveal the associations between the closing values of Solana, Bitcoin, Ethereum, NASDAQ 100, S&P 500, Gold and Brent Oil, and to provide results that will help investors develop effective portfolio diversification strategies and reduce portfolio risk.
Design/methodology/approach – The hypothesis of the study is to examine whether the prices of Solana, Bitcoin, Ethereum, NASDAQ 100, S&P 500, Gold and Brent Oil move together. The data set consists of the closing values of Solana, Bitcoin, Ethereum, NASDAQ 100, S&P 500, Gold and Brent Oil for weekdays and 744 business days when the stock exchanges were open between 04.01.2021 and 19.01.2024. Apriori and FP-Growth association analyses, which are machine learning algorithms, were applied at different support levels to determine associations between variables. Associations among variables were detected at different support levels (0,10 ve 0,25).
Results – According to the results of the Apriori and FP-Growth association analysis, it was observed that the closing values of Bitcoin, Ethereum, Solana, NASDAQ 100 and S&P 500 generally move together. Gold and Brent Oil, on the other hand, have generally exhibited opposite and rarely parallel associations with other variables. Especially in the association rules at the 0.25 support level, no associations with cryptocurrencies have been detected. This is an important finding for investors to reduce portfolio risk.
Discussion – The study examines the assosiations among cryptocurrencies, indices, and commodities, presenting findings that can assist investors in optimizing their portfolio diversification strategies and, consequently, reducing risks more effectively. Moreover, by constructing a portfolio comprising non-correlated assets such as Gold, Brent Crude Oil, and cryptocurrencies, it offers investors opportunities for profit generation and risk reduction, that is, the possibility of portfolio diversification.
Downloads
Published
How to Cite
Issue
Section
License

This work is licensed under a Creative Commons Attribution-NoDerivatives 4.0 International License.