Functional Clustering of Longitudinal Data

  • Jeng-Min Chiou
  • Pai-Ling Li
Part of the Contributions to Statistics book series (CONTRIB.STAT.)

This study considers two clustering criteria to achieve difierent goals of grouping similar curves. These criteria are based on the minimal L 2 distance and the maximal functional correlation defined in this study, respectively. Each cluster centers on a subspace spanned by the cluster mean and covariance eigenfunctions of the underlying random functions. Clusters can thus be identified by the subspace projection of curves.


Random Function Cluster Membership Functional Correlation Cluster Subspace Cluster Criterion 


Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.


  1. [1]
    Bock, H. H.: On the interface between cluster analysis, principal component analysis, and multidimensional scaling. In: H. Bozdogan, A. K. Gupta (eds): Multivariate statistical modeling and data analysis. D. Reidel, Dordrecht, 17-44 (1987).Google Scholar
  2. [2]
    Chiou, J.-M. and Li, P.-L.: Functional clustering and identifying substructures of longitudinal data. J. R. Statist. Soc. B. 69, 679-699 (2007).CrossRefMathSciNetGoogle Scholar
  3. [3]
    García-Escudero, L. A. and Gordaliza, A.: A proposal for robust curve clustering. J. Class. 22, 185-201 (2005).CrossRefGoogle Scholar
  4. [4]
    Ramsay, J. O. and Silverman, B. W.: Functional Data Analysis. 2nd ed. Springer: New York. (2005).Google Scholar
  5. [5]
    Tarpey, T. and Kinateder, K.K.J.: Clustering functional data. J. Class. 20, 93-114 (2003). MATHCrossRefMathSciNetGoogle Scholar

Copyright information

© Physica-Verlag Heidelberg 2008

Authors and Affiliations

  • Jeng-Min Chiou
    • 1
  • Pai-Ling Li
    • 2
  1. 1.Institute of Statistical ScienceAcademia SinicaTaiwan
  2. 2.Department of StatisticsTamkang UniversityTaiwan

Personalised recommendations