Semiparametric Inference And Bandwidth Choice Under Long Memory: Experimental Evidence

Authors : Uwe HASSLER, Maya OLİVARES

Pages : 27-41

View : 18 | Download : 7

Publication Date : 2013-01-01

Article Type : Research Paper

Abstract :The most widely used semiparametric estimators under fractional integration are variants of the  local Whittle [LW] estimator. They are consistent for the long memory parameter d and follow a limiting  normal distribution. Such properties require the bandwidth m to satisfy certain restrictions for the estimators  to be “local” or semiparametric in large samples. Optimal rates for m are known and data-driven selection  procedures have been proposed. A Monte Carlo study is conducted to compare the performance of the LW  and the so-called exact LW estimators both in terms of experimental size when testing hypotheses about d  and in terms of root mean squared error. In particular, the choice of the bandwidth is addressed. Further,  competing approximations to limiting normality are compared.
Keywords : Fractional integration, approximate normality, bandwidth selection