Advertisement

Clustering Upper Level Units in Multilevel Models for Ordinal Data

  • Leonardo GrilliEmail author
  • Agnese Panzera
  • Carla Rampichini
Conference paper
Part of the Studies in Classification, Data Analysis, and Knowledge Organization book series (STUDIES CLASS)

Abstract

We consider an explorative method for unsupervised clustering of upper level units in a two-level hierarchical setting. The idea lies in applying a density-based clustering algorithm to the predicted random effects obtained from a multilevel cumulative logit model. We illustrate the proposed approach throughout the analysis of data from European Social Survey about political trust in European countries.

Keywords

Density-based clustering Empirical Bayes predictions European Social Survey Proportional odds model Random effects 

References

  1. 1.
    Aitkin, M.: A general maximum likelihood analysis of variance components in generalized linear models. Biometrics 55, 117–128 (1999)MathSciNetCrossRefGoogle Scholar
  2. 2.
    Azzimonti, L., Ieva, F., Paganoni, A.M.: Nonlinear nonparametric mixed-effects models for unsupervised classification. Comput. Stat. 28, 1549–1570 (2013)MathSciNetCrossRefGoogle Scholar
  3. 3.
    Asparouhov, T., Muthén, B.: Multilevel mixture models. In: Hancock, G.R. and Samuelsen, K.M. (Eds.) Advances in latent variable mixture models, pp. 27–51, Charlotte, NC: Infurmation Age Publishing (2008)Google Scholar
  4. 4.
    Azzalini, A., Menardi, G.: Clustering via nonparametric density estimation: the R Package pdfCluster. J. Stat. Software 57, 1–26 (2014)CrossRefGoogle Scholar
  5. 5.
    Azzalini, A., Torelli, N.: Clustering via nonparametric density estimation. Statist. Computing 17, 71–80 (2007)MathSciNetCrossRefGoogle Scholar
  6. 6.
    Grilli, L., Rampichini, C.: Multilevel models for ordinal data. In: Kenett, R. and Salini, S. (eds.) Modern Analysis of Customer Surveys: with Applications using R., Chapter 19, Wiley (2012)Google Scholar
  7. 7.
    Heinzl, F., Tutz, G.: Clustering in linear mixed models with approximate Dirichlet process mixtures using EM algorithm. Stat. Modell. 13, 41–67 (2013)MathSciNetCrossRefGoogle Scholar
  8. 8.
    Lukociene, O., Varriale, R., Vermunt, J.K.: The simultaneous decision(s) about the number of lower- and higher-level classes in multilevel latent class analysis. Sociological Meth. 40, 247–283 (2010)CrossRefGoogle Scholar
  9. 9.
    Menardi, G., Azzalini, A.: An advancement in clustering via nonparametric density estimation. Statist. Computing 124, 753–767 (2014)MathSciNetCrossRefGoogle Scholar
  10. 10.
    McCulloch, C.E., Neuhaus, J.M.: Prediction of random effects in linear and generalized linear models under model misspecification. Biometrics 67, 270–279 (2011)MathSciNetCrossRefGoogle Scholar
  11. 11.
    McCulloch, M.E., Neuhaus, J.M.: Misspecifying the shape of a random effects distribution: why getting it wrong may not matter. Stat. Sci. 26, 388–402 (2011)MathSciNetCrossRefGoogle Scholar
  12. 12.
    Nylund, K.L., Asparouhov, T., Muthén, B.O.: Deciding the number of classes in latent class analysis and growth mixture modeling: a Monte Carlo simulation study. Struct. Equat. Model. 14, 535–569 (2007)MathSciNetCrossRefGoogle Scholar
  13. 13.
    Ohlssen, D.I., Sharples, L.D., Spiegelhalter, D.J.: Flexible random-effects models using Bayesian semi-parametric models: applications to institutional comparisons. Statist. Med. 26, 2088–2112 (2007)MathSciNetCrossRefGoogle Scholar
  14. 14.
    Skrondal, A., Rabe-Hesketh, S.: Prediction in multilevel generalized linear models. J. R. Stat. Soc. Ser. A 172, 659–687 (2009)MathSciNetCrossRefGoogle Scholar
  15. 15.
    Vermunt, J.K.: Mixture models for multilevel data sets. In: Hox, J., Roberts, J.K. (eds.) Handbook of Advanced Multilevel Analysis, pp. 59–81. Routledge (2010)Google Scholar
  16. 16.
    Wand, M.P., Jones, M.C.: Kernel smoothing. Chapman & Hall, London (1995)CrossRefGoogle Scholar

Copyright information

© Springer International Publishing AG 2018

Authors and Affiliations

  • Leonardo Grilli
    • 1
    Email author
  • Agnese Panzera
    • 1
  • Carla Rampichini
    • 1
  1. 1.Department of Statistics, Computer ScienceApplications ‘G. Parenti’ University of FlorenceFlorenceItaly

Personalised recommendations