|
Organizers |
A Nonparametric Test for Assessing Spectral Peaks
by
Scott Holan
University of Missouri - Columbia
Coauthors: Tucker McElroy - U.S. Census Bureau
Peaks in the spectrum of a stationary process are indicitive of the presence of a periodic phenomenon, such as a seasonal effect or business cycle. This work proposes to measure and test for the presence of such spectral peaks via assessing their aggregate acceleration and velocity. Our method is developed nonparametrically and thus may be useful in the preliminary analysis of a series. The technique is also useful for detecting the presence of residual seasonality in seasonally adjusted data. The diagnostic is investigated through simulation and several seasonal series.
Date received: October 14, 2005
Copyright © 2005 by the author(s). The author(s) of this document and the organizers of the conference have granted their consent to include this abstract in Atlas Conferences Inc. Document # carm-80.