Macro carbon price prediction with support vector regression and Paris accord targets

TL;DR

This paper presents a support vector regression model that predicts the EU carbon market price in 2030 using economic, political, and market factors, including Paris Accord targets, to aid market decision-making.

Contribution

It introduces a novel macro-level prediction model incorporating political targets like the Paris Accord into carbon price forecasting.

Findings

Support vector regression effectively predicts future carbon prices.

Inclusion of Paris Accord targets improves prediction accuracy.

Model demonstrates potential for practical application in carbon market management.

Abstract

Carbon neutralization is an urgent task in society because of the global warming threat. And carbon trading is an essential market mechanics to solve carbon reduction targets. Macro carbon price prediction is vital in the useful management and decision-making of the carbon market. We focus on the EU carbon market and we choose oil price, coal price, gas price, and DAX index to be the four market factors in predicting carbon price, and also we select carbon emission targets from Paris Accord as the political factor in the carbon market in terms of the macro view of the carbon price prediction. Thus we use these five factors as inputs to predict the future carbon yearly price in 2030 with the support vector regression models. We use grid search and cross validation to guarantee the prediction performance of our models. We believe this model will have great applications in the macro carbon price prediction.