@article {10.3844/jcssp.2025.1705.1718,
article_type = {journal},
title = {Predicting CO2 Emissions Using Multivariate Regression: Assessing the Impact of Pandemics and Industrial Revolutions},
author = {Mekki, Youssef and Moujahdi, Chouaib and Assad, Noureddine and Dahbi, Aziz},
volume = {21},
number = {7},
year = {2025},
month = {Jul},
pages = {1705-1718},
doi = {10.3844/jcssp.2025.1705.1718},
url = {https://thescipub.com/abstract/jcssp.2025.1705.1718},
abstract = {This paper explores the prediction of CO2 emissions usingMultivariate Regression models, incorporating the influence of significanthistorical events such as pandemics and industrial revolutions. Whileexisting research primarily focuses on carbon emissions alone in forecastingmodels, this study emphasizes the importance of incorporating multiplefactors for a comprehensive understanding of CO2 dynamics. Beyondemissions, factors such as socioeconomic indicators, industrial activities,environmental policies, and historical events play significant roles. Thismultidimensional methodology is essential for developing robust predictionmodels capable of capturing the complex dynamics of CO2 emissions. Thispaper introduces the notion of regression flags to mark these impactfulevents, revealing the intricate relationship between CO2 emissions andexternal factors. By integrating these flags into a Multivariate Regressionmodel, we uncover how different historical contexts shape emissions trendsover time. Experimental results underscore the effectiveness of our model inaccurately predicting CO2 emissions dynamics amidst varying historicalconditions. The experimental results demonstrate a robust ability of the used Multivariate Regression model with the notion of events flags to achieveprecise predictions of CO2 emissions over time.},
journal = {Journal of Computer Science},
publisher = {Science Publications}
}