Detection of outliers in gas emissions from urban areas using functional data analysis.

In this work a solution for the problem of the detection of outliers in gas emissions in urban areas that uses functional data analysis is described. Different methodologies for outlier identification have been applied in air pollution studies, with gas emissions considered as vectors whose components are gas concentration values for each observation made. In our methodology we consider gas emissions over time as curves, with outliers obtained by a comparison of curves instead of vectors. The methodology, which is based on the concept of functional depth, was applied to the detection of outliers in gas omissions in the city of Oviedo and results were compared with those obtained using a conventional method based on a comparison of vectors. Finally, the advantages of the functional method are reported.