EFFICIENT DETRENDING IN COINTEGRATING REGRESSION

This paper studies efficient detrending in cointegraling regression and develops modified tests for cointegration that use efficient deirending procedures. Asymptotics for these tests arc derived. Monle Carlo experiments arc conducted lo evaluate the deirending procedures in finite samples and lo compare tests for cointegration based on different detrending procedures. The limit theory allows for increasingly remote initial condition effects as the sample size goes to infinity.