A New Approach to Predicting Cryptocurrency Returns Based on the Gold Prices with Support Vector Machines during the...

Area of Science:

• Quantitative Finance
• Computational Economics
• Financial Technology (FinTech)

Background:

• Accurate cryptocurrency return prediction is challenging, especially during the COVID-19 pandemic.
• Traditional forecasting methods may not suffice for volatile digital asset markets.
• Understanding cryptocurrency dynamics is crucial for investors during uncertain economic periods.

Purpose of the Study:

• To propose a novel approach for predicting cryptocurrency returns relative to gold price quantiles.
• To evaluate the efficacy of Support Vector Machine (SVM) for cryptocurrency return predictability.
• To analyze financial return predictability across pre-COVID-19 and COVID-19 periods.

Main Methods:

• Utilized Support Vector Machine (SVM) algorithm for classification of cryptocurrency returns.
• Selected six major digital currencies: Binance Coin, Bitcoin, Cardano, Dogecoin, Ethereum, and Ripple.
• Incorporated sensor-based data collection and an updated data analysis algorithm.

Main Results:

• Demonstrated strong evidence of SVM's robustness in predicting cryptocurrency returns.
• SVM provided accurate results for profitable trading strategies both before and during the pandemic.
• The proposed algorithm facilitates updated data analysis via sensors.

Conclusions:

• Support Vector Machine (SVM) is a reliable technique for cryptocurrency return prediction and trading strategy development.
• Findings are valuable for stakeholders seeking to understand cryptocurrency market behavior, particularly during crises.
• The study offers insights for improved investment decision-making in volatile financial environments.