Spatially Modeling the Effects of Meteorological Drivers of PM2.5 in the Eastern United States via a Local Linear Penalized Quantile Regression Estimator.

摘要: Fine particulate matter (PM2.5) poses a significant risk to human health, with long-term exposure being linked conditions such as asthma, chronic bronchitis, lung cancer, and atherosclerosis. In order improve the current pollution control strategies better shape public policy, development of more comprehensive understanding this air pollutant is necessary. To end, work attempts quantify relationship between certain meteorological drivers levels PM2.5. It expected that set important will vary both spatially within conditional distribution PM2.5 levels. account for these characteristics, new local linear penalized quantile regression methodology developed. The proposed estimator uniquely selects at every spatial location each performance illustrated through simulation, it then used determine association several over Eastern United States. This analysis suggests primary throughout much States tend differ based on season geographic location, similarities existing “typical” “high”