Proceedings of the 3rd International Conference on Public Management and Big Data Analysis, PMBDA 2023, December 15–17, 2023, Nanjing, China

Research Article

Carbon Market Price Forecasting in China Using Probability Density Recurrent Networks

Download24 downloads
  • @INPROCEEDINGS{10.4108/eai.15-12-2023.2345396,
        author={Zijie  Wei and Yongfei  Hu and Pengyu  Wang and Zhixin  Han and Qianjiao  Xie},
        title={Carbon Market Price Forecasting in China Using Probability Density Recurrent Networks},
        proceedings={Proceedings of the 3rd International Conference on Public Management and Big Data Analysis, PMBDA 2023, December 15--17, 2023, Nanjing, China},
        publisher={EAI},
        proceedings_a={PMBDA},
        year={2024},
        month={5},
        keywords={time series; probability density recurrent networks; carbon price forecasting},
        doi={10.4108/eai.15-12-2023.2345396}
    }
    
  • Zijie Wei
    Yongfei Hu
    Pengyu Wang
    Zhixin Han
    Qianjiao Xie
    Year: 2024
    Carbon Market Price Forecasting in China Using Probability Density Recurrent Networks
    PMBDA
    EAI
    DOI: 10.4108/eai.15-12-2023.2345396
Zijie Wei1, Yongfei Hu1, Pengyu Wang2, Zhixin Han2, Qianjiao Xie2,*
  • 1: Longyuan(Beijing) Carbon Asset Management Technology Co., Ltd.
  • 2: Wuhan University
*Contact email: xieqj1996@whu.edu.cn

Abstract

With the nationwide proliferation of carbon markets, the scarcity of carbon emissions allowances has become comparable to other commodities in traditional markets, endowing them with unique market value. Carbon prices play a central role in carbon market mechanisms, reflecting the outcomes of market competition and shaping government policies related to carbon emissions allocation. Consequently, the accurate prediction of carbon prices is essential for businesses to comprehend fluctuations in carbon prices, efficiently manage carbon emissions, and provide a sound foundation for trading decisions. In this study, we address data uncertainty by developing a probability density recurrent network. We investigate the intricate interrelationships and evolving patterns among data points using probability density distribution functions in time series data, enabling the creation of a prediction model to anticipate future data points. Multiple evaluation criteria corroborate the precision and effectiveness of the methodology employed in this research.