Affiliation:
1. MIMAR SINAN FINE ARTS UNIVERSITY
Abstract
In recent years, Bitcoin (BTC) has become the most popular digital asset in the cryptocurrency market. Its prices are highly volatile due to rapidly increasing investor interest, making it difficult to predict price movements. The aim of this study is to predict trend reversals in BTC price movements by using tree-based ensemble machine learning techniques and compare the success rates of these techniques. For this purpose, the study focuses on points where the trend changes. The ‘buy’, ‘sell’, and ‘hold’ classes are balanced through under-sampling. Extreme Gradient Boosting (XGB), Random Forest (RF) and Random Trees (RT) models are developed. The results are evaluated by using precision, recall, specificity, F1 score and accuracy metrics. The study concludes that the XGB model exhibits higher success compared to other models.
Publisher
Turkish Journal of Forecasting
Reference23 articles.
1. [1] S. Ürgenç, Predicting Bitcoin Trends Reversals With Machine Learning Methods (Makine Öğrenmesi Yöntemleri ile Bitcoin Trend Dönüşlerinin Tahmin Edilmesi), (2023). Master Thesis, Mimar Sinan Fine Arts University, Istanbul.
2. [2] N.T. İnce, Predicting The Bitcoin Trend Using Technical Indicators For Deep Learning Algorithmic Features, (2019). Master Thesis, Boğaziçi University, Istanbul.
3. [3] Z. Qiang, J. Shen, Bitcoin High-Frequency Trend Prediction with Convolutional and Recurrent Neural Networks, Comput. Sci. (2021).
4. [4] S. Cavalli, M. Amoretti, CNN-based multivariate data analysis for bitcoin trend prediction, Appl. Soft Comput. 101 (2021) 107065. doi: 10.1016/J.ASOC.2020.107065.
5. [5] S. Alonso-Monsalve, A.L. Suárez-Cetrulo, A. Cervantes, D. Quintana, Convolution on neural networks for high-frequency trend prediction of cryptocurrency exchange rates using technical indicators, Expert Syst. Appl. 149 (2020) 113250. doi: 10.1016/J.ESWA.2020.113250.