ROBUSTIFYING MODEL-FITTING

A robust method for fitting a model from a parametric family is fundam ental to effective statistical analysis. A procedure is given for robu stifying any model fitting process by using weights from the family fr om which the model is to be chosen. The weighting reduces the influenc e of information that is not compatible with the model family, while m aintaining the basic structure of a familiar model fitting process. Th e procedure produces a parametric family of model fitting functions. T he value of the parameter determines the degree to which the weighting influences the robustified model fit. A mechanism for determining an appropriate value for the parameter is described.