Simulating Mixtures of Multivariate and Regression Data in FSDA with MIXSIM

• Activity Number: 638
• Type: Topic Contributed
• Date/Time: Thursday, August 13, 2015 : 8:30 AM to 10:20 AM
• Sponsor: Section on Statistical Computing
• Author(s): Domenico Perrotta* and Marco Riani and Andrea Cerioli and Francesca Torti
• Companies: EC Joint Research Centre and University of Parma and University of Parma and EC Joint Research Centre
• Keywords: Clustering ; Mixture Models ; Syntetic data ; simulation experiments

Abstract

MixSim is a framework which is useful for evaluating the performance of clustering algorithms on the basis of measures of agreement between data partitioning and flexible generation methods for data. The peculiarity of the method is that data are simulated from normal mixture distributions on the basis of pre-specified synthesis statistics on an overlap measure, defined as sum of pairwise misclassification probabilities. We provide new tools which enable to control additional overlapping statistics and departures from homogeneity and sphericity among groups. The output of this extension is a more flexible generation of data to better address modern robust clustering scenarios. We also study the properties and the implications that this new way of simulating clustering data entails in terms of coverage of space, goodness of fit to theoretical distributions, and degree of convergence to nominal values. We demonstrate the new features using our MATLAB implementation that we have integrated in our FSDA toolbox for MATLAB. With MixSim, FSDA now integrates in the same environment state of the art robust clustering algorithms and principled routines for their evaluation and calibration.