In this paper, a methodology for estimating a regression model with locally stationary errors is proposed. In particular, we consider models that have two features: time-varying trends and errors belonging to a class of locally stationary processes. The proposed procedure provides an efficient...