Exposure model for estimating health effects of air pollution

点击翻译按钮获取中文摘要

For estimating long-term exposure to air pollutants from an ambient air pollution monitoring station network, a set of statistical models was examined in this study, which sequentially estimates outdoor and indoor concentrations, and individual exposure. A hierarchical combination of regression and kriging models enabled to predict outdoor concentration from measured values of monitoring stations with smaller root-mean-square error to actual measurements than other conventional approaches. Adjustment with questionnaire about indoor environment drastically improved the precision of estimation from outdoor to indoor concentration.