Forecasting green bond volatility via novel heterogeneous ensemble approaches

作者：

Highlights：

• We pay a special attention on volatility forecasting of green bond.

• Two novel heterogeneous ensemble models, EX-MV and EX-SEL, are proposed.

• The proposed approaches combine tree-based ensemble models and exogenous predictors.

• The incorporation of exogenous predictors boosts predictive accuracy.

• EX-SEL significantly outperforms the benchmarks in most cases.

摘要

•We pay a special attention on volatility forecasting of green bond.•Two novel heterogeneous ensemble models, EX-MV and EX-SEL, are proposed.•The proposed approaches combine tree-based ensemble models and exogenous predictors.•The incorporation of exogenous predictors boosts predictive accuracy.•EX-SEL significantly outperforms the benchmarks in most cases.

论文关键词：Volatility forecasting,Green bonds,Heterogeneous ensemble,Random forests,Gradient boosting decision tree

论文评审过程：Received 9 June 2021, Revised 5 April 2022, Accepted 10 May 2022, Available online 13 May 2022, Version of Record 29 May 2022.

论文官网地址：https://doi.org/10.1016/j.eswa.2022.117580