Outlier recognition in the crystal structure least-squares modelling by diagnostic techniques based on leverage analysis

摘要

The identification of the actual outliers in a least-squares crystal-structure modelrefinement and their subsequent elimination from the data set is a non-trivialtask that has to be carried out carefully when a high level of accuracy of theestimates is required. One of the most suitable tools for detecting the influenceof each data entry on the regression is the identification of ‘leverage points’. Onthe other hand, the recognition of the actual statistical outliers is effectivelypossible by using some diagnostics as a function of the leverage, such as Cook’sdistance, DFFITS and FVARATIO. The evaluation of these estimators makes itpossible to achieve a reliable identification of the outliers and the elimination ofthose that impair the least-squares fit. In this paper, a procedure for filteringdata points based on this kind of analysis for crystallographic X-ray data ispresented and discussed.