The Excess Mass Approach and the Analysis of Multi-Modality

  • G. Sawitzki
Part of the Studies in Classification, Data Analysis, and Knowledge Organization book series (STUDIES CLASS)


The excess mass approach is a general approach to statistical analysis. It can be used to formulate a probabilistic model for clustering and can be applied to the analysis of multi-modality. Intuitively, a mode is present where an excess of probability mass is concentrated. This intuitive idea can be formalized directly by means of the excess mass functional. There is no need for intervening steps like initial density estimation. The excess mass measures the local difference of a given distribution to a reference model, usually the uniform distribution. The excess mass defines a functional which can be estimated efficiently from the data and can be used to test for multi-modality.


Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.


  1. DAVIES, L. (1994): Data features. Manuscript. Essen 1994. To appear in Statistica Nederlandica. Google Scholar
  2. DONOHO, D.L. (1988): One-sided inference about functionais of a density. The Annals of Statistics, 16, 1390–1420CrossRefGoogle Scholar
  3. GORDON, A.D. (1994): Null models in cluster vaiiadation. In: W. Gaul, D. Pfeifer (eds.) From data to knowledge: Theoretical and practical aspects of classification, data analysis and knowledge organization. Proc. 18th Annual Conference of the GfKl, Univ. of Oldenburg, 1994. Springer Verlag, Heidelberg Berlin, 1994.Google Scholar
  4. HARTIGAN, J.A. (1975): Clustering algorithms. Wiley, NewYork.Google Scholar
  5. HARTIGAN, J.A., and H ARTIGAN, P.M. (1985): The dip test of unimodality. Annals of Statistics, 13 70–84.CrossRefGoogle Scholar
  6. HARTIGAN, J.A. (1987): Estimation of a convex density contour in two dimensions. Journal of the American Statistical Association, 82, 267–270.CrossRefGoogle Scholar
  7. MÜLLER, D.W., and SAWITZKI, G. (1987): Using excess mass estimates to investigate the modality of a distribution. Preprint Nr. 398, Januar 1987, Universität Heidelberg, Sonderforschungsbereich 123 Stochastische Mathematische Modelle.Google Scholar
  8. MÜLLER, D.W., and SAWITZKI, G. (1991): Excess mass estimates and tests for multimodaiity.Journal of the American Statistical Association, 86 738–746.CrossRefGoogle Scholar
  9. MÜLLER, D.W. (1992): The excess mass approach in statistics. Beiträge zur Statistik 3. ftp: Google Scholar
  10. NASON, G.P., and SIBSON, R. (1992): Measuring multimodaiity. Statistics and Computing 2, 153–160.CrossRefGoogle Scholar
  11. NOLAN, D. (1991): The excess-mass ellipsoid. Journal of Multivariate Analysis, 39 348–371.CrossRefGoogle Scholar
  12. POLONIK, W. (1993): Measuring mass concentration and estimating density cöntour clusters - an excess mass approach. Beiträge zur Statistik 7. ftp: statlab.uni- Submitted to Annals of Statistics. Google Scholar
  13. POLONIK, W. (1993a): Density estimation under qualitative assumptions in higher dimensions. Beiträge zur Statistik 15. ftp: Google Scholar

Copyright information

© Springer-Verlag Berlin · Heidelberg 1996

Authors and Affiliations

  • G. Sawitzki
    • 1
  1. 1.StatLab HeidelbergHeidelbergGermany

Personalised recommendations