Extreme Learning Machine-Autoregression Integrated Moving Average Composite Model

Abstract:

Excessive carbon dioxide emissions are the primary factor causing global warming. Currently, models for controlling carbon dioxide emissions mainly focus on population, economy, and technology. A significant amount of research has been conducted on multivariate linear regression analysis encompassing factors such as population, GDP, and energy consumption. However, the studies examining the impact of green finance on emissions have been limited to qualitative and semi-quantitative levels, lacking in-depth and systematic research. This study establishes a composite model combining the Autoregressive Integrated Moving Average (ARIMA) model and Extreme Learning Machine (ELM) model. This composite model is employed to analyze the impact of Hubei provinces’ permanent resident population, GDP, comprehensive energy consumption, and green finance index on carbon dioxide emissions. In the ELM model, the impact of four variables-population, GDP, comprehensive energy consumption, and green finance index-on carbon dioxide emissions is investigated. Using data from 2002 to 2019, the ARIMA model is applied to predict these four variables after differencing. The ELM model's prediction of carbon dioxide emissions has a very small relative error compared to actual results. The composite mode- ELM-ARIMA model is used to analyze the province's carbon dioxide emissions from 2024 to 2030.

• Periodical: Advances in Science and Technology (Volume 166)
• Pages: 133-139
• DOI: https://doi.org/10.4028/p-4rJyXE
• Online since: June 2025
• Authors: Zi Wen Yin, Chun Jiang*
• Keywords: Autoregressive Integrated Moving Average, CO₂ Emission, Extreme Learning Machine, Green Finance Index

References

[1] Yang, Y., Ingwersen, W. W., Hawkins, T. R., Srocka, M., and Meyer, D. E., USEEIO: A New and Transparent United Stated Environmentally-Extended Input-Output Model, Journal of Cleaner Production, 158, 308, 2017.

DOI: 10.1016/j.jclepro.2017.04.150

[2] Daniel Isakovic, et al, GIS APPLICATIONS IN AIR POLLUTION MODELING, 2022 International Conference on Computational Science and Computational Intelligence (CSCI), 2022, pp.1315-1321.

DOI: 10.1109/csci58124.2022.00235

[3] Ghoddusi, H., Creamer, G. G., Rafizadeh, N. Machine learning in energy economics and finance: A review. Energy Economics, 81:709–727, 2019.

DOI: 10.1016/j.eneco.2019.05.006

[4] Wei Sun, Caifei Wang, Chongchong Zhang, Factor analysis and forecasting of CO2 emissions in Hebei, using extreme learning machine based on particle swarm optimization, Journal of Cleaner Production, Vol.162, 20 September 2017, pp.1095-1101.

DOI: 10.1016/j.jclepro.2017.06.016

[5] N. Assessing the impact of green finance and urbanization on the tourism industry—an empirical study in china. Environmental Science and Pollution Research, 30:3576–3592, 2023.

DOI: 10.1007/s11356-022-22207-5

[6] Amato, F., Guignard, F., Robert, S. et al. A novel framework for spatio-temporal prediction of environmental data using deep learning. Sci Rep 10, 22243 (2020).

DOI: 10.1038/s41598-020-79148-7

[7] Patrick Canning,Sarah Rehkamp, Jing Yi, Environmental Input-Output (EIO) Models for Food Systems Research: Application and Extensions, Food Systems Modelling, 2022, pp.179-212.

DOI: 10.1016/b978-0-12-822112-9.00014-x

[8] Cai Meng,Ren Chao. GIS-Based Modeling for Estimating Urban Carbon Emissions, CRC Press, 2024. EBook ISBN: 9781003244561.

DOI: 10.1201/9781003244561-17