Search academic texts

Information × Registration Number 2123U006674, Article popup.category Препринт Title popup.author Shkredko Andriy popup.publication 01-01-2023 popup.source_user Український католицький університет popup.source https://hdl.handle.net/20.500.14570/4808 popup.publisher Description This study aims to assess the performance of diverse linear and non-linear machine learning models in predicting cryptocurrency returns. Additionally, we examine the variation in model metrics across different prediction intervals to gain insights into the impact of temporal factors on the accuracy and explainability of volatility pat- terns. Furthermore, the study explores the influence of normalization on model per- formance, particularly when incorporating new features. Special attention is given to analyzing the effect of Elon Musk’s tweets on cryptocurrency returns and the in- clusion of sentiment-based tweet features, focusing on evaluating the implications for model accuracy and explanatory volatility capabilities. popup.nrat_date 2025-05-09 Close