ABSTRACT
The partially nonlinear model has been popular in the statistical literature since they retain the flexibility of nonparametric models and the easy interpretation of parametric models. In this chapter, a new estimation procedure is developed for estimating the parameters of the partially nonlinear models. For this purpose, penalized profile nonlinear least square problem where nonparametric part of the partially nonlinear model is expressed as an additive model in terms of B-spline basis function. Then, the estimation problem is handled by considering kernel function in bridge penalty method. The solution of the generated problems will be obtained with Conic Quadratic Programming (CQP) as one of the methods of convex optimization, and that solution will be called C-Bridge Estimator.
