Nonparametric estimation for some nonlinear models

Godambe's (1985) theorem on optimal estimating equations for stochastic processes is applied to nonparametric estimation problems for nonlinear time-series models with time-varying parameter [alpha](t). Examples are considered from the usual classes of nonlinear time-series models. The goal of this paper is to arrive at a nonparametric estimate of [theta]0 = [alpha](t0) for a fixed point t0 [epsilon] [0, 1].Nonlinear Nonparametric Estimation Estimating function Autoregressive Random coefficient Kernel