We consider nonparametric estimation of the regression function g(*) in a nonlinear regression model Y<sub>t</sub> = g(X<sub>t</sub>) o(X<sub>t</sub>)e<sub>t</sub>, where the regressor X<sub>t</sub> is a nonstationary unit root process and the error e<sub>t</sub> is s sequence of independent and identically distributed (i.i.d.) random variables. With proper...
