Sparse convex combinations of forecasting models by meta learning

作者：

Highlights：

• Sparse forecast combinations can be obtained within a meta-learning framework.

• Combining a subset of forecasting methods does not affect the predictive accuracy.

• Sparsity of a forecast combination is attractive for real-time forecasting systems.

• Forecasts of TBATS and ARIMA models are almost never excluded from the combination.

摘要

•Sparse forecast combinations can be obtained within a meta-learning framework.•Combining a subset of forecasting methods does not affect the predictive accuracy.•Sparsity of a forecast combination is attractive for real-time forecasting systems.•Forecasts of TBATS and ARIMA models are almost never excluded from the combination.

论文关键词：Forecasting,Forecast combination,Meta-learning,Sparsity,Real-time-systems

论文评审过程：Received 19 October 2021, Revised 22 January 2022, Accepted 17 March 2022, Available online 28 March 2022, Version of Record 4 April 2022.

论文官网地址：https://doi.org/10.1016/j.eswa.2022.116938