A1035
Title: Local Whittle estimation in time varying long memory series
Authors: Josu Arteche - University of the Basque Country (Spain) [presenting]
Abstract: The memory parameter is usually assumed to be constant in traditional long memory time series. We relax this restriction by considering the memory of the series to be a function that depends on a finite number of parameters, allowing it to vary smoothly with time. A Local Whittle-type estimator of these parameters is proposed. Its consistency and asymptotic normality are shown for locally stationary and non-stationary long memory processes, where the spectral behaviour is only locally restricted around the origin. Its good finite sample performance is shown in a Monte Carlo and two empirical applications, supporting its benefits over the fully parametric Whittle estimator.
