Linear Regression with PM2.5 and PM10 Concentration for Air Quality in East Lima, Peru

This research paper provides an in-depth analysis of the impact of meteorological factors on the regression of PM2.5 and PM10 concentrations and their correlation with SARS-COV-2 transmission in Peru, employing heteroscedasticity considerations. The findings of the study reveal significant OLS estimators and intercepts for both PM2.5 and PM10, with remarkable comparisons with previous research underlining the robustness of the developed models. In particular, regression models designed for different settings (rural, urban and industrial) highlight how PM10 concentrations are predictive of PM2.5 levels, with coefficients of determination exceeding those found in previous studies. In addition, the article uses the Kruskal-Wallis test and the Holm-Bonferroni method to demonstrate significant variations in particulate matter in different study areas, lending further credence to the comprehensive nature of the analysis performed. Given the rigorous quantitative analysis, the clear significance of the findings in multiple settings, and the contribution of the study to the understanding of the complex relationship between air pollution and virus transmission, this article is a valuable addition to the field. The detailed comparison with existing models, along with the innovative use of statistical methods to validate the results, positions this research as a significant advance in environmental science and public health. A regression equation between PM2.5 and PM10 was determined; in rural area with linear coefficient of 0.66105 \(\upmu {\text{g}}/{\text{m}}^{3}\) and intercept of 5.57229 \(\upmu {\text{g}}/{\text{m}}^{3}\) , for urban it has a coefficient of 0. 672768 \(\upmu {\text{g}}/{\text{m}}^{3}\) and intercept of 1.149877 \(\upmu {\text{g}}/{\text{m}}^{3}\) , and industrial a coefficient of 0.567541 \(\upmu {\text{g}}/{\text{m}}^{3}\) and an intercept of 4.485924.