Bitcoin Price Prediction: A Machine Learning Approach Using Opening and Closing Data

Dornala, Sahithi (2024) Bitcoin Price Prediction: A Machine Learning Approach Using Opening and Closing Data. Masters thesis, Dublin, National College of Ireland.

Preview	PDF (Master of Science) Download (828kB) | Preview
Preview	PDF (Configuration Manual) Download (821kB) | Preview

Abstract

This study concentrates on predicting and forecasting the opening and closing prices of Bitcoin utilizing enhanced machine learning models. The dataset used is the “Bitcoin Historical Data,” which contains Bitcoin price data recorded every minute with related Open, High, Low, and Close (OHLC) prices and Bitcoin and USD trading volumes beginning from 2015. The first goal was to examine the patterns of Bitcoin’s market movement and create a reliable predictive model for its prices. The models used in this study include GRU (Gated Recurrent Units), LSTM (Long Short-Term Memory), RNN (Recurrent Neural Networks), Prophet, and Prophet-GRU. Based on the results, while ProphetGRU was hypothesized to leverage the strengths of both models, it did not outperform the standalone GRU model. Specifically, the GRU model demonstrated superior performance, with an R² Score of 0.9959, an RMSE of 0.0041, and an MAE of 0.0027 for the Opening Price, and an R² Score of 0.9959, an RMSE of 0.0026, and an MAE of 0.0042 for the Closing Price. In contrast, Prophet-GRU exhibited lower accuracy with higher error rates (e.g., RMSE of 18.661 and 40.561 for Open and Close Prices, respectively). These findings highlight GRU’s strength in high-frequency financial data forecasting, whereas Prophet struggled to add value due to the lack of significant seasonality in the dataset. This study provides insights into the integration of machine learning and statistical models for Bitcoin price forecasting and serves as a useful resource for traders and analysts within the cryptocurrency market.

Item Type:	Thesis (Masters)
Supervisors:	Name Email Rustam, Furqan UNSPECIFIED
Uncontrolled Keywords:	Bitcoin Forecasting; Cryptocurrency Price Prediction; GRU (Gated Recurrent Units); Time Series Analysis; Opening Price Prediction; Closing Price Prediction
Subjects:	Q Science > QA Mathematics > Electronic computers. Computer science T Technology > T Technology (General) > Information Technology > Electronic computers. Computer science H Social Sciences > HG Finance > Money > Digital currency > Cryptocurrencies Q Science > Q Science (General) > Self-organizing systems. Conscious automata > Machine learning
Divisions:	School of Computing > Master of Science in Data Analytics
Depositing User:	Ciara O'Brien
Date Deposited:	02 Sep 2025 10:22
Last Modified:	02 Sep 2025 10:22
URI:	https://norma.ncirl.ie/id/eprint/8695

Actions (login required)

View Item