|
Abstract:
|
Multivariate receptor modeling (MRM) is a collection of methods used for identifying major pollution sources and estimating their impacts from concentration data of air pollutants. Typical MRM deals with source identification and their contribution for the mean concentration, and it can be called as mean MRM. Air pollution data are often highly right-skewed with outliers of large values due to high level air pollution. For such highly skewed data, the median (50th percentile) rather than the mean may be more appropriate as a parameter representing the center of data. Moreover, in some cases the modeling of very high concentrations of air pollutants such as 95th percentiles and estimating the corresponding source contributions and compositions, which could be different from those for the mean concentrations, may be of specific interest, e.g., in setting the standards for pollution control. In this study, we propose a Bayesian quantile MRM which can be applied to any quantile of concentration data, in a Bayesian framework. The proposed method is illustrated with simulated data and PM2.5 speciation data from El Paso, Texas, USA.
|
