International Carbon Market Price Forecasting Research Based on ARIMA-RF Model

The carbon market price forecasting may help policy makers adopt effective measures to maintain carbon market order, and help financial institutions and enterprises avoid investment risks. However, the research of carbon price forecasting is inherently highly complex and traditional linear time series analysis models are difficult to achieve satisfactory results In this paper, an integration model based on ARIMA-RF is proposed to predict international carbon price. The model includes two steps: the original sequence of carbon price is divided into linear and nonlinear parts at the first phase, and then ARIMA and random foresets are respectively used to fit the linear and nonlinear subsequence. Additionally, we can obtain the final prediction result by integrating two-parts.