Inference on Nonstationary Time Series with Moving Mean

A semiparametric model is proposed in which a parametric filtering of a non-stationary time series, incorporating fractionally differencing with short memory correction, removes correlation but leaves a nonparametric deterministic trend. Estimates of the memory parameter and other dependence parameters are proposed, and shown to be consistent and asymptotically normally distributed with parametric rate. Unit root tests with standard asymptotics are thereby justified. Estimation of the trend function is also considered. We include a Monte Carlo study of finite-sample performance.