Research On Influencing Factors And Prediction Of Carbon Trading Market Price In China

Global climate change,which has become the greatest threat to human sustainable development,is drawing increased attention from the most international society.In response to climate change,a series of actions have been taken internationally.The most effective action for global warming is the formulation and implementation of the Kyoto Protocol,which aims to reduce carbon emissions in the atmosphere within a certain period of time.In order to achieve this goal,the carbon emission trading market mechanism has been established,which is the most effective tool to mitigate greenhouse gas emissions such as carbon dioxide in the world.As the largest greenhouse gas emitter in the world,as well as the largest energy and coal consumer in the world,China’s economic and social development is facing tremendous pressure on energy conservation and emission reduction.To this end,the Chinese government launched the pilot project of carbon emission trading in China in October 2011,and in full-scale launched the national carbon emissions trading system in 2017.Considering that carbon price is the core of the carbon trading market and an important basis for the government to formulate policies and enterprises to make emission reduction decisions,this paper studies and analyzes the influencing factors of carbon trading price in China,and based on this,predicts the future price of carbon trading market.This paper divides the influencing factors of carbon trading market price into extrinsic influencing factors and intrinsic influencing factors.The extrinsic influencing factors refer to the impact of external factors such as foreign carbon price,energy price,macro economy,etc.on carbon prices.The intrinsic influencing factors refer to the influence of the historical price of carbon trading market on the current price.Firstly,five external factors affecting the price of China’s carbon trading market are analyzed from the theoretical level: foreign carbon price,energy price,macro economy,climate environment and exchange rate changes.Then,taking the Guangdong carbon price GDEA as an example,the vector autoregressive(VAR)model and the factor augmented vector autoregressive(FAVAR)model are used to quantitatively analyze the relationship between the extrinsic influencing factors and the price of China’s carbon trading market.The results show that the impact of various extrinsic influencing factors on the Guangdong carbon price GDEA is delayed by one period;among all the indicators,the energy price factor has the strongest impact,followed by the foreign macroeconomic factors and the daily average temperature in Guangzhou has the weakest impact and can be ignored;the impact of CER futures prices is higher than the impact of EUA futures prices;the impact of Euro-RMB Exchange Rate in the first two periods is small,and until the third period is rapidly increasing.With regard to the intrinsic influencing factors,this paper also analyzes both theoretical and empirical aspects,the study found that the price of carbon trading market is greatly affected by its pre-price.Therefore,the intrinsic influencing factors of carbon trading market price cannot be ignored when conducting carbon price forecasting.For the sake of obtaining a more reliable carbon price forecasting method,this paper proposes a combined model of fast ensemble empirical mode decomposition algorithm(FEEMD)and extreme learning machine optimized by particle swarm optimization(PSO-ELM),in which the FEEMD algorithm is utilized to decompose carbon price for noise reduction purposes,the PSO-ELM is employed to predict the components that are decomposed.Based on the analysis results of carbon price influencing factors and the carbon price forecasting model,the empirical analysis is carried out on the data of carbon price and its influencing factors of the carbon trading market in Guangdong Province.The partial autocorrelation function(PACF)is exploited to select the intrinsic influencing factors,which are lagging sequences that have a large correlation with the carbon price series.The principal component analysis(PCA)is used to find out the extrinsic influencing factors,which are five principal components obtained by dimension reduction of 38 selected influencing factors.And then the intrinsic and extrinsic influencing factors are combined as inputs to the forecasting model PSO-ELM.The prediction results demonstrate that the selected influencing factors are reasonable,and incorporating them into the constructed model for carbon price prediction can achieve higher fitting accuracy.This conclusion can provide more effective and reliable information for policy makers and market participants,and enrich and expand the research content of carbon trading market price,which has certain guiding significance for the development of China’s carbon trading market.