CMStatistics 2022: Start Registration
View Submission - CMStatistics
B0431
Title: Understanding the spillover effects of the air pollution mixture using mobility data Authors:  Joseph Antonelli - University of Florida (United States) [presenting]
Abstract: Estimating the causal effects of air pollution is an important problem as we require a better understanding of the nature of this relationship in order to guide future regulation. Of particular interest is the impact of air pollution mixtures, i.e. the joint impact of multiple air pollutants on health. Many cohort studies assign pollution levels to individuals based on their home zip code, but individuals travel to multiple zip codes with potentially different pollution levels. To provide a better understanding of the overall impact of the air pollution mixture, we target the effects of both exposure to pollution within an individual's zip code and exposure to pollution from other zip codes. We use a weighted average of exposure to pollution from other zip codes by incorporating cell phone mobility data. Using nonparametric Bayesian models, we then estimate the spatial spillover effect of air pollution exposure.