Simultaneous semi-parametric estimation of clustering and regression

We investigate the parameter estimation of regression models with fixed group effects, when variable is missing while group-related variables are available. This problem involves clustering to infer based on variables, and build a model target given eventually some additional variables. Thus, this can be formulated as joint distribution modeling The usual strategy for two-step approach starting by learning (clustering step) then plugging in its estimator fitting (regression step). However, suboptimal (providing particular biased estimates) since it does not make use clustering. we advise simultaneous both regression, semiparametric framework. Numerical experiments illustrate benefits our proposition considering wide ranges distributions models. relevance new method illustrated real data dealing problems associated high blood pressure prevention. proposed implemented R package ClusPred available CRAN. Supplementary materials containing technical details codes online.