Journal of Classification

, Volume 11, Issue 1, pp 5–36

The MAP test for multimodality

  • Gregory Paul M. Rozál
  • J. A. Hartigan

Abstract

We introduce a test for detecting multimodality in distributions based on minimal constrained spanning trees. We define a Minimal Ascending Path Spanning Tree (MAPST) on a set of points as a spanning tree that has the minimal possible sum of lengths of links with the constraint that starting from any link, the lengths of the links are non-increasing towards a root node. We define similarly MAPSTs with more than one root. We present some algorithms for finding such trees. Based on these trees, we devise a test for multimodality, called the MAP Test (for Minimal Ascending Path). Using simulations, we estimate percentage points of the MAP statistic and assess the power of the test. Finally, we illustrate the use of MAPSTs for determining the number of modes in a distribution of positions of galaxies on photographic plates from a rich galaxy cluster.

Keywords

Minimal constrained spanning trees Nearest neighbor density estimates Minimal ascending path spanning trees Tests for modes The MAP test 

Preview

Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.

References

  1. DRESSLER, A. (1980), “A Catalog of Morphological Types in 55 Rich Clusters of Galaxies,”Astrophysical Journal Supplement Series, 42, 565–609.Google Scholar
  2. GIACOMELLI, F., WIENER, J., KRUSKAL, J. B., POMERANZ, J. W., and LOUD, A. V. (1971), “Subpopulations of Blood Lymphocytes as Demonstrated by Quantitative Cytochemistry,”Journal of Histochemistry and Cytochemistry, 19, 426–433.Google Scholar
  3. HARTIGAN, J. A., and HARTIGAN, P. M. (1985), “The Dip Test of Unimodality,”Annals of Statistics, 13, 70–84.Google Scholar
  4. HARTIGAN, J. A. (1988), “The Span Test for Unimodality,” inClassification and Related Methods of Data Analysis, Ed. H. H. Bock, Amsterdam: North-Holland, 229–236.Google Scholar
  5. HARTIGAN, J.A. and MOHANTY, S. (1992), “The RUNT Test for Multimodality,”Journal of Classification, 9, 63–70.Google Scholar
  6. PRIM, R. C. (1957), “Shortest Connection Networks and some Generalizations,”Bell System Technical Journal, 36, 1389–1401.Google Scholar
  7. SILVERMAN, B. W. (1981), “Using Kernel Density Estimates to Investigate Multimodality,”Journal of the Royal Statistical Society, Series B, 43, 97–99.Google Scholar
  8. SILVERMAN, B. W. (1986),Density Estimation for Statistics and Data Analysis, New York: Chapman and Hall.Google Scholar
  9. ZZAHN, C. T. (1971), “Graph-Theoretical Methods for Detecting and Describing Gestalt Clusters,”IEEE Transactions on Computers, C-20, 68–86.Google Scholar

Copyright information

© Springer-Verlag 1994

Authors and Affiliations

  • Gregory Paul M. Rozál
    • 1
  • J. A. Hartigan
    • 1
  1. 1.Department of StatisticsYale UniversityNew Haven

Personalised recommendations