A New N-factor Affine Term Structure Model of Futures Price for CO2 Emissions Allowances: Empirical Evidence from the EU ETS

In recent years, carbon emission markets have become liquid and promising markets within the European Union emissions trading scheme (EU ETS). In order to fit and forecast futures price for CO2 emissions allowances, we propose a new N-factor affine term structure model for CO2 futures price and estimate parameters in the new affine model by using the Kalman filter technique. Our empirical results show that CO2 futures price follow significant mean-reversion process, and the estimated coefficients of meanreversion speed, market risk premium, volatility and correlation among state variables are almost significant. Compared with one-factor model, mean absolute errors (MAE) and root mean square errors (RMSE) in prediction errors from two-factor and three-factor model are lower, accordingly two-factor and three-factor model can accurately describe the term structure of CO2 futures price. Key-Words: CO2 emissions allowances; futures prices; new affine model; term structure; Kalman filter