Frequency Domain Bootstrap

  • S. N. Lahiri
Part of the Springer Series in Statistics book series (SSS)


In this chapter, we describe a special type of transformation based-bootstrap, known as the frequency domain bootstrap (FDB). Given a finite stretch of observations from a stationary time series, here we consider the discrete Fourier transforms (DFTs) of the data and use the transformed values in the frequency domain to derive bootstrap approximations (hence, the name FDB). In Section 9.2, we describe the FDB for a class of estimators, called the ratio statistics. Dahlhaus and Janas’s (1996) results show that under suitable regularity conditions, the FDB is second-order accurate for approximating the sampling distributions of ratio statistics. In Section 9.3, we describe the FDB method and its properties in the context of spectral density estimation. Material covered in Section 9.3 is based on the work of Franke and Härdle (1992). In Section 9.4, we describe a modified version of the FDB due to Kreiss and Paparoditis (2003) that, under suitable regularity conditions, removes some of the limitations of the standard FDB and yields valid approximations to the distributions of a larger class of statistics than the class of ratio statistics. It is worth pointing out that the results presented in this chapter on the FDB are valid only for linear processes.


Spectral Density Ratio Statistic Linear Process Valid Approximation Optimal Bandwidth 
These keywords were added by machine and not by the authors. This process is experimental and the keywords may be updated as the learning algorithm improves.


Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.

Copyright information

© Springer Science+Business Media New York 2003

Authors and Affiliations

  • S. N. Lahiri
    • 1
  1. 1.Department of StatisticsIowa State UniversityAmesUSA

Personalised recommendations