Analysis of Bitcoin Price Prediction Using Machine Learning

Abstract

The research purpose of this paper is to obtain an algorithm model with high prediction accuracy for the price of Bitcoin on the next day through random forest regression and LSTM, and to explain which variables have influence on the price of Bitcoin. There is much prior literature on Bitcoin price prediction research, and the research methods mainly revolve around the ARMA model of time series and the LSTM algorithm of deep learning. Although it cannot be proved by the Diebold–Mariano test that the prediction accuracy of random forest regression is significantly better than that of LSTM, the prediction errors RMSE and MAPE of random forest regression are better than those of LSTM. The changes in the variables that determine the price of Bitcoin in each period are also obtained through random forest regression. From 2015 to 2018, three US stock market indexes, NASDAQ, DJI, and S&P500 and oil price, and ETH price have impact on Bitcoin prices. Since 2018, the important variables have become ETH price and Japanese stock market index JP225. The relationship between accuracy and the number of periods of explanatory variables brought into the model shows that for predicting the price of Bitcoin for the next day, the model with only one lag of the explanatory variables has the best prediction accuracy.

Bitcoin volatility in bull vs. bear market-insights from analyzing on-chain metrics and Twitter posts

“…When comparing the results of the PA with Chen (2023) , the PA achieves superior results in both root mean squared error (RMSE) and MAPE, as shown in Table 7 . The improvement in performance is attributed to the enhanced data collection, data analysis and processing, and model selection and hyperparameter tuning steps.…”

“…The authors concluded that on-chain metrics are a good supplementary tool to existing deep learning techniques when it comes to cryptocurrency price prediction. Most recently, Chen (2023) developed a Bitcoin price prediction system using an RF model and 47 metrics, divided into eight categories: Bitcoin price variables, technical features of Bitcoin, other cryptocurrencies, commodities, market index, foreign exchange, public attention, and dummy variables of the week. The model was trained on two periods, one ranging from April 2015 to October 2018 and the other from October 2018 to April 2022.…”

Exploring the relationship between digital trails of social signals and bitcoin returns

“…The primary stream of research aims at discovering the dynamics of the cryptocurrency market and its potential influence on the global financial system (Dwyer, 2015; Hendrickson et al , 2016; Fama et al , 2019; Morillon, 2022; Shakri, 2022). Other popular sub-streams aim at predicting bitcoin price movements (Morillon and Chacon, 2022; Chen, 2023) and comprehending the volatility of bitcoin price returns (Aalborg et al , 2019; Miglietti et al , 2019; Kinateder and Papavassiliou, 2021; Sosa et al , 2022). Another widely explored sub-stream of research in the domain is examining netizens’ thoughts and expectations regarding bitcoin price returns (e.g.…”

Research on the Stock Price Prediction Using Machine Learning

“…An algorithm model combined random forest regression and LSTM was chosen to predict the price of Bitcoin by Junwei Chen [4].…”