Study On Combined Forecasting Of Overcapacity In China's Coal Industry And Its Governance Strategy

After experiencing the rapid development of the golden decade,China's coal industry is facing a situation of oversupply.The problem of coal overcapacity has become increasingly prominent and needs to be resolved.To this end,the central government has formulated a series of de-capacity policies in recent years.However,the formation mechanism of coal overcapacity is complicated and has various forms of performance.In the process of de-capacity,multiple stakeholders are involved,which makes the task of de-capacity face many challenges.In order to achieve accurate prediction and effective management of coal overcapacity,on the basis of combing the relevant literature on overcapacity,this paper focused on the exploratory and empirical research on the development trend,the structural characteristics of causes and the governance strategy of overcapacity in China's coal industry.Firstly,in order to quantitatively describe the structural characteristics and development trend of coal overcapacity in China,based on the idea of “decompositionreconstruction-prediction”,a combined forecasting method and model(EEMDLSSVM-ARIMA)is proposed,which include a methods set of empirical mode decomposition(EEMD),least squares support vector machine(LSSVM)and the autoregressive integral moving average model(ARIMA).The EEMD method is used to decompose the time series of coal overcapacity scale,and some IMF items and RES items with different time scale are obtained.These items are constructed into high frequency component(HFC),low frequency component(LFC)and trend component(RES)by the fine-to-coarse method,the economic meanings are identified according to their volatility characteristics.Finally,using the LSSVM method that is good at nonlinear series to predict HFC and LFC,and RES is predicted by the ARIMA method that is suitable for linear series.The prediction results of each sub-component are summed to obtain the final prediction result.Secondly,considering the current governance predicament of overcapacity in China's coal industry,this paper discusses the overcapacity governance strategy relying on evolutionary game theory.A threeparty evolutionary game model including the central and local government and coal enterprises is established.The behavioral interaction mechanism among the three parties is discussed from the perspective of dynamic evolution.The impact of different parameter changes on the system and the stability of each game part is studied.Finally,based on the result of combined forecasting and evolutionary game analysis,this paper proposes some policy recommendations for overcapacity management in the coal industry from the perspective of different stakeholders,which cover the macro level,the meso level and the micro level.The study found that:(1)In view of the nonlinear and non-stationary characteristics of the time series of coal overcapacity,this paper constructs a combined forecasting model(EEMD-LSSVM-ARIMA)based on data characteristics.The empirical results and comparative analysis show that,compared with the existing single prediction model(ARIMA,LSSVM)and dual hybrid prediction model(EEMDARIMA,EEMD-LSSVM),the model of EEMD-LSSVM-ARIMA has better prediction performance in terms of horizontal accuracy and vertical accuracy.(2)Using the established model to predict the development trend of coal overcapacity.The result shows that,during the period of 2017~2019,the overcapacity scale in China's coal industry is between 1.7 billion tons and 1.8 billion tons,whose fluctuation is relatively stable.But the overcapacity in China's coal industry still has a long-term surplus.(3)Among the main driving forces of coal overcapacity,the impact of institutional distortion is the most important factor compared to economic cyclical fluctuations and market failures,whose influence degree is around 70%.(4)Based on certain assumptions,the strategy combination of the system finally reaching a stable state is that the central government chooses a market-oriented regulation strategy,local governments choose to strictly implement the strategy,and coal enterprises also choose a positive response strategy.This proves that market-oriented regulation is an effective mechanism to solve the problem of overcapacity in China' coal industry.(5)In the process of overcapacity management in the coal industry,the strategic choices of the single entity of the central government,local governments and coal enterprises will be affected by the strategy of other's,indicating that coal de-capacity should coordinate the interests of multiple entities,which provides a theoretical basis for making overcapacity governance strategies.(6)Controlling the de-capacity cost,constructing a performance appraisal system that emphasizes both economic development and ecological quality,appropriate penalty and restriction for de-capacity policy implementation,increasing the investment level in technological innovation and industrial upgrading play an important role in promoting the effective governance of overcapacity in China's coal industry.In general,the study results of this paper have important theoretical and practical value.In theory,it improves the forecasting method of industrial overcapacity and enriches the theory of coal overcapacity governance.In practice,it provides a theoretical guidance for the government and coal enterprises to accurately grasp the laws,causes and evolution trends of overcapacity in China's coal industry,also provides a certain strategic support for the coal overcapacity management.