Macroeconomic forecasting through news, emotions and narrative

作者：

Highlights：

• Presentation of effective filtering methodology that isolates relevant signals.

• Emotions derived from global newspapers have predictive power for macroeconomic variables.

• Of seven distinct emotion groups, “happiness” and “anger” have the strongest predictive power.

摘要

•Presentation of effective filtering methodology that isolates relevant signals.•Emotions derived from global newspapers have predictive power for macroeconomic variables.•Of seven distinct emotion groups, “happiness” and “anger” have the strongest predictive power.

论文关键词：News sentiment,Time series forecasting,Big data,Natural language processing

论文评审过程：Received 29 September 2020, Revised 17 February 2021, Accepted 17 February 2021, Available online 25 February 2021, Version of Record 21 March 2021.

论文官网地址：https://doi.org/10.1016/j.eswa.2021.114760