Download PDF
Methods Inf Med 1990; 29(03): 236-242
DOI: 10.1055/s-0038-1634785
Statistical Analysis
Schattauer GmbH

Detection of Aberrant Observations in a Background of an Unknown Multidimensional Gaussian Distribution

E. S. Gelsema
1   Department of Medical Informatics, Erasmus University Rotterdam, The Netherlands
,
B. Leijnse
2   Department of Chemical Pathology, Faculty of Medicine and Health Sciences and Academic Hospital Dijkzigt, Erasmus University Rotterdam, The Netherlands
,
R. W. Wulkan
2   Department of Chemical Pathology, Faculty of Medicine and Health Sciences and Academic Hospital Dijkzigt, Erasmus University Rotterdam, The Netherlands
› Author Affiliations
Further Information

Publication History

Publication Date:
07 February 2018 (online)

    Permissions and Reprints

Abstract

An exploratory iterative technique for the detection of aberrant observations on a background of a multidimensional Gaussian distribution is described. Its development was motivated by the analysis of a set of three measurements reflecting the acid-base metabolism in the blood of 2,402 intensive care patients. This new, three-dimensional treatment of such data yields a meaningful description. A technical evaluation of the method, using artificially generated data is also presented. It is shown that the model parameters of the underlying Gaussian distributions are determined with good accuracy and that the accuracy with which the contamination is estimated increases with increasing distance of the contaminating observations from the mean.

Key-Words

Normality Assessment - Outlier Detection - Acid-Base Metabolism
 

  • REFERENCES

  • 1 Albert A, Harris EK. Multivariate Interpretation of Clinical Laboratory Data. New York: Marcel Dekker Inc; 1987
    Google Scholar
  • 2 D’Agostino RB, Stephens MA. Goodness of Fit Techniques. New York: Marcel Dekker Inc; 1986
    Google Scholar
  • 3 Andrews DF, Gnanadesikan R, Warner JL. Methods for assessing multivariate normality In: Krishnaiah PR. ed. Proceedings of the International Symposium on Multivariate Analysis 1973; 03: 95-116.
    PubMed
  • 4 Wilks SS. Multivariate statistical outliers. Sankhya 1963; A25, 407-26.
    PubMedGoogle Scholar
  • 5 Gnanadesikan R, Kettenring JR. Robust estimates, residuals and outlier detection with multiresponse data. Biometrics 1972; 28: 81-124.
    CrossrefPubMedGoogle Scholar
  • 6 Wilk MB, Gnanadesikan R. Probability plotting methods for the analysis of data. Biometrika 1968; 55: 1-17.
    PubMedGoogle Scholar
  • 7 Davenport HW. ABC of Acid Base Chemistry. Chicago: University of Chicago Press; 1977
    Google Scholar
  • 8 Krusinska E, Liebhart J. The influence of outliers on discrimination of chronic obturative lung disease. Meth Inform Med 1988; 27: 167-76.
    PubMedGoogle Scholar
  • 9 Mahalanobis PC. On the generalized distance in statistics. Proc Nat Inst Sci 1936; 12: 49-55.
    PubMedGoogle Scholar
  • 10 Day NE. Estimating the components of a mixture of normal distributions. Biometrika 1969; 56: 463-71.
    CrossrefPubMedGoogle Scholar
  • 11 Chan TF, Golub GH, LeVeque RJ. Updating formulae and a pairwise algorithm for computing sample variances. In: Caus-sinus H, Ettinger P, Tomassone R. eds. COMPSTAT 1982, Proceedings in Computational Statistics. Vienna: Physica Ver-lag; 1982: 130-41.
    Google Scholar