Abstract:
|
Quality control methods for observations for which more than one variable is recorded are generally based on using robust estimates of parameters for a particular distribution, and that particular distribution is usually the multivariate normal (MVN). However, many multivariate data generating processes do not produce elliptical contours, and in such cases, error detection using the MVN distribution would lead to many legitimate observations being erroneously flagged. In this work, we propose non-parametric and parametric methods for identifying errors in skewed bivariate data. In the first method, we remove potential outliers by assigning each bivariate observation a depth score and remove those observations that fall beyond a given threshold. In the second method, we first develop robust estimators for the parameters in a bivariate skew-t (BST) distribution, and these parameters are used in either a distance-based or contour-based approach to flag observations as potential outliers. We test the performance of these methods in simulation against a more common MVN outlier detection method and apply them to radiosonde launches of the wind vector for 8 pressure levels.
|
ASA Meetings Department
732 North Washington Street, Alexandria, VA 22314
(703) 684-1221 • meetings@amstat.org
Copyright © American Statistical Association.