Outlier Detection with Kernel Density Functions in Monitoring the ISTAT LFS Data Production Processes


To calculate the thresholds that determine the normal operating region for a data production process, it is proposed a method based on a time-varying tolerance bounds. The proposal is applied on the Italian Labour Force Survey to report how this new decision rules impact on survey quality

DOI Code: 10.1285/i20705948v6n1p118

Keywords: monitoring data production, AAPOR, time series analysis


Ornella GIAMBALVO, Statistical methods for data analysis,


Luciana QUATTROCIOCCHI, Survey monitoring luciana.quattrociocchi@istat.it

Full Text: PDF

Creative Commons License
This work is licensed under a Creative Commons Attribuzione - Non commerciale - Non opere derivate 3.0 Italia License.