Please use this identifier to cite or link to this item: https://hdl.handle.net/10419/62857 
Year of Publication: 
2006
Series/Report no.: 
Working Paper No. 570
Publisher: 
Queen Mary University of London, Department of Economics, London
Abstract: 
This paper considers estimation and inference in some general non linear time series models which are embedded in a strongly dependent, long memory process. Some new results are provided on the properties of a time domain MLE for these models. The paper also includes a detailed simulation study which compares the time domain MLE with a two step estimator, where the Local Whittle estimator has been initially employed to filter out the long memory component. The time domain MLE is found to be generally superior to two step estimation. Further, the simulation study documents the difficulty of precisely estimating the parameter associated with the speed of transition. Finally, the fractionally integrated, nonlinear autoregressive-ESTAR model is found to be extremely useful in representing some financial time series such as the forward premium and real exchange rates.
Subjects: 
Non-linearity, ESTAR models, Strong dependence, Forward premium, Real exchange rates
JEL: 
C22
C12
F31
Document Type: 
Working Paper

Files in This Item:
File
Size
402.78 kB





Items in EconStor are protected by copyright, with all rights reserved, unless otherwise indicated.