Research Seminar
Clusterwise simultaneous component analysisKim De RooverKU Leuven | |
| Abstract: | Numerous research questions in educational sciences and psychology concern the structure of a set of variables. To study the structure of a set of variables, one typically relies on scores from a group of persons on those variables. One may, however, wonder whether the same structure would have been retrieved if another group of persons had been studied. To trace such structural differences, one will have to gather data from different groups. Formally, the resulting data then constitute multivariate multiblock data, with persons being nested in groups. Obviously, the crucial question is how such data have to be analyzed to find out whether and in what way the structure of the variables differs across the groups of persons. A number of principal component analysis techniques exists to study such structural differences, for instance, simultaneous component analysis (SCA). However, these techniques suffer from some important limitations. Therefore, in this presentation, we propose a novel generic modeling strategy, called Clusterwise SCA-P, which solves these limitations and which encompasses several existing techniques as special cases. Clusterwise SCA-P generalizes the earlier proposed Clusterwise SCA-ECP model (De Roover et al., 2010). Like Clusterwise SCA-ECP, Clusterwise SCA-P partitions the groups into a number of clusters and specifies a simultaneous component model for each cluster. Unlike Clusterwise SCA-ECP, Clusterwise SCA-P allows for between-group differences in the variances and the correlations of the cluster-specific components. As such, the Clusterwise SCA-P model offers a more detailed insight into group differences. An algorithm for fitting Clusterwise SCA-P solutions is presented and its performance is evaluated by means of a simulation study. The value of the model for empirical research is illustrated with data from the domain of psychiatric diagnosis. |
| Date: | Tue Oct 12, 12:15 pm - 1:15 pm |
| Place: | room 03.60 (Department of Psychology, Tiensestraat 102, 3000 Leuven) |