Search

Cluster-Weighted Modeling (CWM) is a flexible mixture approach for modeling the joint probability of data coming from a heterogeneous population as a weighted sum of the products of marginal distributions and conditional distributions. In this paper, we introduce a wide family of Cluster Weighted models in which the conditional distributions are assumed to belong to the exponential family with canonical links which will be referred to as Generalized Linear Gaussian Cluster Weighted Models. Moreover, we show that, in a suitable sense, mixtures of generalized linear models can be considered as nested in Generalized Linear Gaussian Cluster Weighted Models. The proposal is illustrated through many numerical studies based on both simulated and real data sets.

Cluster-weighted modeling (CWM) is a mixture approach for modeling the joint probability of a response variable and a set of explanatory variables. The parameters are estimated by means of the expectation-maximization algorithm according to the maximum likelihood approach. Under Gaussian assumptions, we analyse the complete-data likelihood function of cluster weighted models. Further, under suitable hypotheses we show that the maximization of the likelihood function of Gaussian cluster weighted models leads to the same parameter estimates of finite mixtures of regression and finite mixtures of regression with concomitant variables. In this sense, the latter ones can be considered as nested models of Gaussian cluster weighted models., Comment: The paper has been withdrawn because the results contained therein have been revised in a more general framework and moved into a more extended paper about generalized cwm

A novel family of twelve mixture models with random covariates, nested in the linear $t$ cluster-weighted model (CWM), is introduced for model-based clustering. The linear $t$ CWM was recently presented as a robust alternative to the better known linear Gaussian CWM. The proposed family of models provides a unified framework that also includes the linear Gaussian CWM as a special case. Maximum likelihood parameter estimation is carried out within the EM framework, and both the BIC and the ICL are used for model selection. A simple and effective hierarchical random initialization is also proposed for the EM algorithm. The novel model-based clustering technique is illustrated in some applications to real data. Finally, a simulation study for evaluating the performance of the BIC and the ICL is presented.

We investigate statistical properties of Cluster-Weighted Modeling, which is a framework for supervised learning originally developed in order to recreate a digital violin with traditional inputs and realistic sound. The analysis is carried out in comparison with Finite Mixtures of Regression models. Based on some geometrical arguments, we highlight that Cluster-WeightedModeling provides a quite general framework for local statistical modeling. Theoretical results are illustrated on the ground of some numerical simulations.