Forecasting cryptocurrencies prices using data driven level set fuzzy models

作者：

Highlights：

• The work develops data-driven fuzzy models to forecast cryptocurrencies prices.

• Modeling is based on level set concerning on how fuzzy rules are built and processed.

• Output functions map the activation levels of fuzzy rules directly in model outputs.

• One-step-ahead forecasting of the closing prices of cryptocurrencies is performed.

• Level set-based fuzzy modeling performs best in terms of direction of price change.

摘要

•The work develops data-driven fuzzy models to forecast cryptocurrencies prices.•Modeling is based on level set concerning on how fuzzy rules are built and processed.•Output functions map the activation levels of fuzzy rules directly in model outputs.•One-step-ahead forecasting of the closing prices of cryptocurrencies is performed.•Level set-based fuzzy modeling performs best in terms of direction of price change.

论文关键词：Data driven fuzzy modeling,Cryptocurrency,Forecasting

论文评审过程：Received 9 May 2022, Revised 15 July 2022, Accepted 1 August 2022, Available online 6 August 2022, Version of Record 12 August 2022.

论文官网地址：https://doi.org/10.1016/j.eswa.2022.118387