مخلوط های محدود برازش رگرسیون خطی عمومی در R

R package flexmix provides flexible modelling of finite mixtures of regression models using the EM algorithm. Several new features of the software such as fixed and nested varying effects for mixtures of generalized linear models and multinomial regression for a priori probabilities given concomitant variables are introduced. The use of the software in addition to model selection is demonstrated on a logistic regression example.

Finite mixtures of regression models are a popular method to model unobserved heterogeneity or to account for overdispersion in data. They are flexible models and in theory it is easy to modify and extend them by using more complex models for the component distribution functions and estimate the corresponding parameters, e.g., using the EM algorithm. R (R Development Core Team, 2006) features several extension packages for estimation of mixture regression models, e.g., fpc for mixtures of linear regression models (Hennig, 2000) and mmlcr for mixed-mode latent class regression (Buyske, 2006). However, like virtually all other (non-RR) implementations, they consider only a few particular types of mixture models and do not reflect the generality of the theoretical model class in the software design. RR package flexmix (Leisch, 2004) tries to fill this gap by encapsulating the abstract statistical objects of interest into S4 classes and methods such that the resulting software can be easily extended. This paper is organized as follows: Section 2 gives notation and the model class, the main new functions of flexmix are presented in Section 3, and we end with a short demonstration in Section 4. The latest development version of the package sources and all RR code necessary to reproduce the results in this article are available from http://www.ci.tuwien.ac.at/research/mixtures.