A one-step forward forecasting test for carbon market pricing is done in this work, in which data from the first seven trading days is used to anticipate the price on the eighth trading day. The study compares the MAE, MSE, and RMSE values of several forecasting models and discovers that combining empirical mode decomposition (EMD) and forecasting models yields the best results. It was shown that the hybrid model can improve both the durability and accuracy of carbon price estimates. In terms of forecasting, the combined EMD-BiLSTM-ATTENTION model beats other comparator models, and carbon price forecasting errors in Hubei and Fujian are smaller than those in Shenzhen due to their more stable frequency amplitudes. Nevertheless, it has been discovered that estimating the carbon price in Shenzhen is more difficult due to higher amplitude variations, resulting in higher prediction errors. Overall, the findings indicate that the proposed EMD-BiLSTM-ATTENTION model is appropriate for carbon price prediction, and the study includes carbon market price prediction maps for Shenzhen, Hubei, and Fujian.


    Access

    Check access

    Check availability in my library

    Order at Subito €


    Export, share and cite



    Title :

    Carbon Price Prediction based on EMD-BiLSTM-ATTENTION model


    Contributors:
    Li, Tong (author) / Li, Shilun (author) / Lin, Feng (author) / Zhuo, Xingxuan (author)


    Publication date :

    2023-08-25


    Size :

    1568461 byte




    Type of media :

    Conference paper


    Type of material :

    Electronic Resource


    Language :

    English



    Attention-BiLSTM-based ship track prediction method

    YANG ZE / GUAN ZHENBO / GE XIANG et al. | European Patent Office | 2025

    Free access

    Taxi-Passenger’s Destination Prediction via GPS Embedding and Attention-Based BiLSTM Model

    Liao, Chengwu / Chen, Chao / Xiang, Chaocan et al. | IEEE | 2022