Lung-Chang Chien (Epidemiology and Biostatistics) recently co-authored a publication titled "" in the Journal of Exposure Science and Environmental Epidemiology. This study addresses the challenge of limited regulatory monitors for PM2.5 in Texas by predicting daily concentrations using machine learning and satellite-derived data. Gradient boosted trees and random forest models were employed, showing strong in-sample performance but varied out-of-sample results. Predicted PM2.5 concentrations generally decrease over time. This research contributes to PM2.5 monitoring and decision-making, emphasizing machine learning's effectiveness in addressing environmental challenges.
