Carbon price prediction considering climate change: A text-based framework

Abstract

Carbon trading is a vital market mechanism to achieve carbon emission reduction. The accurate prediction of the carbon price is conducive to the effective management and decision-making of the carbon trading market. However, existing research on carbon price forecasting has ignored the impacts of multiple factors on the carbon price, especially climate change. This study proposes a text-based framework for carbon price prediction that considers the impact of climate change. Textual online news is innovatively employed to construct a climate-related variable. The information is combined with other variables affecting the carbon price to forecast the carbon price, using a long short-term memory network and random forest model. The results demonstrate that the prediction accuracy of the carbon price in the Hubei and Guangdong carbon markets is enhanced by adding the textual variable that measures climate change.