Tests of long-range dependence in interest rates using wavelets

The fractional parameter d of the percentage change in selected interest rate series is estimated. A methodology that uses properties of wavelets is used to derive an OLS estimator of d. The method uses the ability of wavelets to localize a process in time-scale space. Statistically significant long...

Full description

Saved in:
Bibliographic Details
Published inThe Quarterly review of economics and finance Vol. 44; no. 1; p. 180
Main Authors McCarthy, Joseph, DiSario, Robert, Saraoglu, Hakan, Li, Hsi
Format Journal Article
LanguageEnglish
Published Greenwich Elsevier Science Ltd 01.02.2004
Subjects
Government securities
Interest rates
Mathematical models
Studies
Time series
United States > US
Online AccessGet full text

Cover

More Information
Summary:The fractional parameter d of the percentage change in selected interest rate series is estimated. A methodology that uses properties of wavelets is used to derive an OLS estimator of d. The method uses the ability of wavelets to localize a process in time-scale space. Statistically significant long-range dependence in the percentage changes of interest rates is reported for most of the securities investigated. The methodology tests the hypothesis that the change in the log of the interest rates follows a fractionally integrated process with long-range dependence as characterized by the hyperbolic decay of its autocovariance function and defined by a fractional differencing parameter, 0 < d < 12. The results, with are statistically significant, show that d is in the (0, 0.5) interval for the majority of the data series in the study.
ISSN:1062-9769
DOI:10.1016/S1062-9769(03)00029-2