ERCIM Working Group

Computing & Statistics

Mixture Models

Mixture models experience sustainable popularity over recent years. Not only that they are natural models to adjust for unobserved or latent heterogeneity, they are fundamental cornerstones in many areas in statistics such as smoothing, empirical Bayes, likelihood based clustering, or latent variable analysis among others. As semi-parametric models they combine par excellence the compromise in the trade-off between imposed model structure and freedom in model adaptation to the data. However, mixture models experience a number of difficulties. The likelihood may not be bounded, and, even if it were, the global maximum might not be a good choice. Algorithmic solutions are nearly almost required and algorithms such as the EM algoirthm is experiencing numerous problems such as the choice of initial values or using an adequate stopping rule. The number of components problem and model selection add one more to the many areas of interest. Diverse application areas such as capture-rapture approaches or clustering of gene expression data have been added to numerous existing application areas such as disease mapping or meta-analysis. The track is primarily devoted to these newly emerging issues.


Dankmar Bohning, University of Southampton, UK. E-mail: Send
Marco Alfo, University "La Sapienza", Rome, Italy. E-mail: Send
Valentin Patilea, CREST-ENSAI, France E-mail: Send


    1. Marco Alfo, University "La Sapienza", Rome, Italy
    2. Patrice Bertrand, ENST Bretagne, Rennes, France
    3. Dankmar Bohning, University of Southampton, UK
    4. Christophe Biernacki, Universite Lille 1, France
    5. Daniela G. Calo, University of Bologna, Italy
    6. Gilles Celeux, Universite Paris-Sud, France
    7. Antoine Chambaz, Universite Paris 5 Rene Descartes, France
    8. Gabriela Ciuperca, Universite Claude Bernard Lyon I, France
    9. Alessio Farcomeni, University "Sapienza", Rome, Italy
    10. Anton Forman, University of Vienna, Austria
    11. Herwig Friedl, Graz University of Technology, Austria
    12. Sylvia Fruhwirth-Schnatter, Johannes-Kepler University Linz, Austria
    13. Bernard Garel, Institute National Polytechnique de Toulouse, France
    14. Elisabeth Gassiat, Universite Paris-Sud, Orsay, France
    15. Gerard Govaert, Universite de Technologie Compiegne, France
    16. Bettina Grun, Technische Universitat Wien, Austria
    17. Yann Guedon, CIRAD, Montpellier, France
    18. Christian Hennig, UCL, UK
    19. Heinz Holling, Univeristy of Munster, Germany
    20. Salvatore Ingrassia, Universita di Catania, Italy
    21. Dimitris Karlis, Athens University of Economics, Greece
    22. Christine Keribin, Universite Paris-Sud, Orsay, France
    23. Ronny Kuhnert, Robert-Koch Institute, Berlin, Germany
    24. Friedrich Leisch, Ludwig-Maximilian-Universitat Munchen, Germany
    25. Jean-Michel Marin, Universite Paris-Sud, France
    26. Francesca Martella, University of Leuven, Belgium
    27. Antonello Maruotti, University "Sapienza", Rome, Italy
    28. Luciano Nieddu, Libera Universita degli Studi "San Pio V", Rome, Italy
    29. Valentin Patilea, CREST-ENSAI, France
    30. Denys Pommeret, Universite de la Mediteranee Aix-Marseille II, France
    31. Christian P. Robert, Universite Paris-Dauphine, France
    32. Roberto Rocci, University "Tor Vergata", Rome, Italy
    33. Guillaume Saint-Pierre, INRIA Futurs, France
    34. Peter Schlattmann, Institute for Medical Biometry, Charite Berlin, Germany
    35. Peter Schlattmann, Institute for Medical Biometry, Charite Berlin, Germany
    36. Krunoslav Sever, Helmut Schmidt University, Germany
    37. Peter M. Steiner, Institute for Advanced Studies, Vienna, Austria
    38. Michael Titterington, University of Glasgow, UK
    39. Jeroen Vermunt, Tilburg University, Netherlands
    40. Donatella Vicari, University "La Sapienza", Rome, Italy
    41. Maurizio Vichi, University "La Sapienza", Rome, Italy
    42. Marco Di Zio , National Statistical Institute, Rome, Italy
    43. Victor del Rio Vilas, Veterinary Laboratories Agency, Weybridge, UK

Created by Computing & Statistics Working Group 2007