Open Conference Systems, 50th Scientific meeting of the Italian Statistical Society

Font Size: 
A dissimilarity-based splitting criterion for CUBREMOT
carmela cappelli, Rosaria Simone, Francesca Di Iorio

Last modified: 2018-05-17

Abstract


CUBREMOT  (CUB REgression MOdel Trees) is a model-based approach to grow trees  for ordinal responses that relies on a class of mixture models for evaluations and preferences (CUB). The original proposal considers deviances in log-likelihood to partition observations.  In the present paper a new splitting criterion is introduced that, among the significant splitting variables, chooses the one that maximizes  a dissimilarity measure. This choice is tailored at generating  child nodes as far apart as  possible with respect to the estimated probability distributions. An application to real data on Italians' trust towards the European Parliament taken from the official survey on daily life conducted by the Italian National Institute of Statistics (ISTAT) in 2015 is presented and discussed in comparison with alternative methods.

References


  1. Cappelli, C., Simone, R. and Di Iorio, F. Model-based trees to classify perception and uncer- tainty: analyzing trust in European institutions, under review, 2017.

  2. Cappelli, C., Simone, R., Di Iorio, F. (2017). Growing happiness: a model-based tree. In SIS 2017. Statistics and Data Science: new challenges, new generations. Proceedings of the Con- ference of the Italian Statistical Society, Florence 28–30 June 2017, 261-266.

  3. D’Elia A., Piccolo D. (2005). A mixture model for preference data analysis. Computational Statistics & Data Analysis, 49, 917–934.

  4. GalimbertiG.,SoffrittiG.,DiMasoM.(2012).ClassificationTreesforOrdinalResponsesin R: The RpartScore Package, Journal of Statistical Software, 47, 1–25.

  5. Iannario M., Piccolo D., Simone, R. (2017). CUB: A Class of Mixture Models for Ordinal Data. R package, version 1.1.1 http://CRAN.R-project.org/package=CUB.

  6. Picarretta R. (2008). Classification trees for ordinal variables, Computational Statistics, 23, 407–427.

  7. Leti,G.(1983).Statisticadescrittiva.IlMulino,Bologna.

  8. Zeileis A., Hothorn T., Hornik K. (2008). Model-Based Recursive Partitioning, Journal of Computational and Graphical Statistics, 17, 492–514.

 


Full Text: PDF