Natural gas volatility prediction: Fresh evidence from extreme weather and extended GARCH-MIDAS-ES model

This study aims to analyzes the predictability of the natural gas volatility by considering extreme weather information. Based on extended GARCH-MIDAS models, empirical results show that the predictive model adding weather indicators can indeed outperform the model without weather indicators. Import...

Full description

Saved in:
Bibliographic Details
Published in:Energy economics 2022-12, Vol.116, p.106437, Article 106437
Main Authors: Liang, Chao, Xia, Zhenglan, Lai, Xiaodong, Wang, Lu
Format: Article
Language:English
Subjects:
Extreme weather
GARCH-MIDAS
Natural gas volatility
Volatility forecasting
Citations: Items that this one cites
Items that cite this one
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:This study aims to analyzes the predictability of the natural gas volatility by considering extreme weather information. Based on extended GARCH-MIDAS models, empirical results show that the predictive model adding weather indicators can indeed outperform the model without weather indicators. Importantly, some extreme weather indicators can provide more valuable information to predict the natural gas volatility based on the various out-of-sample tests. Our new weather-related GARCH-MIDAS-ES model can exhibit a new insight on the natural gas volatility forecasting. •We design an extended GARCH-MIDAS models including extreme and normal weather.•The predictive model adding weather indicators can indeed outperform the model without weather indicators.•GARCH-MIDAS-W-ES models including temperature or precipitation achieve the best accuracy forecasts.
ISSN:0140-9883
1873-6181
DOI:10.1016/j.eneco.2022.106437