Complexity analysis and forecasting of variations in cryptocurrency trading volume with support vector regression tuned by Bayesian optimization under different kernels: An empirical comparison from a large dataset

作者：

Highlights：

• SVR under various kernel is employed for cryptocurrency trading volume forecasting.

• Bayesian optimization method is used to tune hyperparameters of SVR.

• ARIMA, Lasso regression and Gaussian regression are considered as benchmark models.

• SVR-BO outperforms all benchmark models.

摘要

•SVR under various kernel is employed for cryptocurrency trading volume forecasting.•Bayesian optimization method is used to tune hyperparameters of SVR.•ARIMA, Lasso regression and Gaussian regression are considered as benchmark models.•SVR-BO outperforms all benchmark models.

论文关键词：Cryptocurrency volume of transactions forecasting,Support vector regression,Bayesian optimization,Forecasting,Hurst exponent,Sample entropy,Largest lyapunov exponent

论文评审过程：Received 29 October 2021, Revised 11 July 2022, Accepted 31 July 2022, Available online 4 August 2022, Version of Record 8 August 2022.

论文官网地址：https://doi.org/10.1016/j.eswa.2022.118349