Forecasting the realized volatility of stock price index: A hybrid model integrating CEEMDAN and LSTM

作者：

Highlights：

• Introduce the hybrid model CEEMDAN-LSTM to forecast RV of stock price index.

• The MCS test is adopted as evaluation criterion of forecast performance.

• Hybrid models with CEEMDAN outperform their corresponding single models.

• CEEMDAN-LSTM performs the best in both emerging and developed markets.

摘要

•Introduce the hybrid model CEEMDAN-LSTM to forecast RV of stock price index.•The MCS test is adopted as evaluation criterion of forecast performance.•Hybrid models with CEEMDAN outperform their corresponding single models.•CEEMDAN-LSTM performs the best in both emerging and developed markets.

论文关键词：Volatility forecasting,Realized volatility,LSTM,CEEMDAN,MCS test

论文评审过程：Received 28 April 2020, Revised 11 January 2021, Accepted 31 May 2022, Available online 3 June 2022, Version of Record 27 June 2022.

论文官网地址：https://doi.org/10.1016/j.eswa.2022.117736