Forecasting mid-price movement of Bitcoin futures using machine learning

Abstract

In the aftermath of the global financial crisis and ongoing COVID-19 pandemic, investors face challenges in understanding price dynamics across assets. This paper explores the performance of the various type of machine learning algorithms (MLAs) to predict mid-price movement for Bitcoin futures prices. We use high-frequency intraday data to evaluate the relative forecasting performances across various time frequencies, ranging between 5 and 60-min. Our findings show that the average classification accuracy for five out of the six MLAs is consistently above the 50% threshold, indicating that MLAs outperform benchmark models such as ARIMA and random walk in forecasting Bitcoin futures prices. This highlights the importance and relevance of MLAs to produce accurate forecasts for bitcoin futures prices during the COVID-19 turmoil.

Forecasting nonperforming loans using machine learning

“…Our conclusion is also in line with earlier studies on machine learning‐based models. For example, Ozgur et al (2021), Bonato et al (2023), and Akyildirim et al (2021) report better output with the random forest model in different forecasting scenarios.…”

“…He documented that machine learning algorithms are adaptable enough to capture systemic reform and time-varying information in a set of predictors to forecast housing returns. Akyildirim et al (2021) used high-frequency intraday data to forecast Bitcoin price movement and showed machine learning algorithms outperform benchmark models such as ARIMA and random walk. Several other studies, that is, Bonato et al (2023), Cepni, Gupta, Pienaar, andPierdzioch (2022), andCepni et al (2019), also documented the implication of machine learning models in financial and macroeconomic time series forecasting.…”

“…In recent years, machine learning‐based models have shown remarkable success in forecasting. Due to their capacity to uncover embedded trends in financial data, machine learning models have also been widely used in various financial applications, such as asset pricing, bankruptcy prediction, high‐frequency trading, and inflation rate forecasting (Abdullah, 2021; Akyildirim et al, 2021; Cepni, Gupta, & Onay, 2022; Gu et al, 2021; Liu et al, 2022; Sulong et al, 2022; Tang et al, 2020; Uddin et al, 2021; Zhang et al, 2022). Following their success, in this paper, we exploit the powerful nonlinear modeling capacity of machine learning to improve the forecasting accuracy of NPLs.…”

The differential influence of social media sentiment on cryptocurrency returns and volatility during COVID-19

“…Investigating the determinants of the returns and volatility of cryptocurrencies has been the focus of a broad number of academic studies ( Dyhrberg et al, 2018 , Eross et al, 2019 , Katsiampa, 2017 , Katsiampa et al, 2019a , Katsiampa et al, 2019b , Akyildirim et al, 2020 Akyildirim et al, 2021 , Papadamou et al, 2021 , Sensoy et al, 2021 ). This research sets out to build on this work, and further investigate whether non-linear causal linkages exist between Twitter-derived measures of economic and market uncertainty and the largest cryptocurrencies during the COVID-19 pandemic.…”

Financial Futures Prediction Using Fuzzy Rough Set and Synthetic Minority Oversampling Technique

“…However, the performance of a trading decision support model will be affected by significant differences in the base price of various stocks [18]. Therefore, their trading decision support systems tend to forecast price movements as a trading signal for the trading strategies.…”