Parametric Versus Semi and Nonparametric Regression Models

There are three common types of regression models: parametric, semiparametric and nonparametric regression. The model should be used to fit the real data depends on how much information is available about form relationship between response variable explanatory variables, random error distribution that assumed. Researchers need familiar with each modeling approach requirements. In this paper, differences these models, estimation methods, robust estimation, applications introduced. For parametric there many known methods such as least squares maximum likelihood which extensively studied but they require strong assumptions. On other hand, models free assumptions regarding response-explanatory variables relationships kernel spline smoothing computationally expensive parameters obtained. two estimators: local constant linear methods. terms bias, especially at boundaries range, better than estimator.  Robust for well studied, however in limited. A method