Cluster-weighted modelling (CWM) is a flexible statistical framework for modelling local relationships in heterogeneous populations on the basis of weighted combinations of local models.We will extend cluster weighted models to include an underlying latent factor structure of the independent variable resulting in a family of parsimonious clusterweighted t-factor analyzers (CWtFA). This provides the model with the flexibility of clustering of high-dimensional data. Expectation-maximization framework along with Bayesian information criterion (BIC) will be used for parameter estimation and model selection. The approach is illustrated on simulated data sets as well as a real data set.
Cluster-weighted t-factor Analyzers for Clustering of High-dimensional Data
PUNZO, ANTONIO;INGRASSIA, Salvatore;
2013-01-01
Abstract
Cluster-weighted modelling (CWM) is a flexible statistical framework for modelling local relationships in heterogeneous populations on the basis of weighted combinations of local models.We will extend cluster weighted models to include an underlying latent factor structure of the independent variable resulting in a family of parsimonious clusterweighted t-factor analyzers (CWtFA). This provides the model with the flexibility of clustering of high-dimensional data. Expectation-maximization framework along with Bayesian information criterion (BIC) will be used for parameter estimation and model selection. The approach is illustrated on simulated data sets as well as a real data set.I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.