Biometrika Advance Access published online on April 24, 2008
Biometrika, doi:10.1093/biomet/asn005
Articles |
Simultaneous confidence bands in spectral density estimation
Institut für Stochastik, Friedrich-Schiller-Universität Jena, Ernst-Abbe-Platz 2, D–07743 Jena, Germany mneumann{at}mathematik.uni-jena.de
Department of Mathematics and Statistics, University of Cyprus, P.O. Box 20537, CY–1678 Nicosia, Cyprus stathisp{at}ucy.ac.cy
Received for publication 1 July 2006. Revision received 1 September 2007.
We propose a method for the construction of simultaneous confidence bands for a smoothed version of the spectral density of a Gaussian process based on nonparametric kernel estimators obtained by smoothing the periodogram. A studentized statistic is used to determine the width of the band at each frequency and a frequency-domain bootstrap approach is employed to estimate the distribution of the supremum of this statistic over all frequencies. We prove by means of strong approximations that the bootstrap estimates consistently the distribution of the supremum deviation of interest and, consequently, that the proposed confidence bands achieve asymptotically the desired simultaneous coverage probability. The behaviour of our method in finite-sample situations is investigated by simulations and a real-life data example demonstrates its applicability in time series analysis.
Key Words: Bootstrap • Confidence band • Gaussian process • Spectral density • Strong approximation
References
-
Beltrão K. L., Bloomfield P. Determining the bandwidth of a kernel spectrum estimate. J. Time Ser. Anal. (1987) 8:21–38.
Brillinger D. R. Time Series. Data Analysis and Theory (1981) New York: McGraw-Hill.
Brockwell P. J., Davis R. A. Time Series: Theory and Methods (1991) 2nd ed. New York: Springer.
Dahlhaus R., Janas D. A frequency domain bootstrap for ratio statistics in time series analysis. Ann. Statist. (1996) 24:1934–63.[CrossRef]
Fan J., Yao Q., Fan J. Nonlinear Time Series: Nonparametric and Parametric Methods (2003) New York: Springer.
Franke J., Härdle W. On bootstrapping kernel spectral estimates. Ann. Statist. (1992) 20:121–45.[CrossRef]
Hall P. Edgeworth expansions for nonparametric density estimators, with applications. Statistics (1991) 22:215–32.[CrossRef]
Hall P. Effect of bias estimation on coverage accuracy of bootstrap confidence intervals for a probability density. Ann. Statist. (1992) 20:675–94.[CrossRef]
Hrafnkelsson B., Newton J. H. Asymptotic simultaneous confidence bands for vector autoregressive spectra. Biometrika (2000) 87:173–82.
Hurvich C. M. Data-driven choice of spectrum estimation: extending the applicability of cross-validation methods. J. Am. Statist. Assoc. (1985) 80:933–40.[CrossRef][ISI]
Koslov J. W., Jones R. H. A unified approach to confidence bounds for the autoregressive spectral estimator. J. Time Ser. Anal. (1985) 6:141–51.
Muller H. G., Prewitt K. Weak convergence and adaptive peak estimation for spectral densities. Ann. Statist. (1992) 20:1329–49.[CrossRef]
Newton J. H., Pagano M. Simultaneous confidence bands for autoregressive spectra. Biometrika (1984) 71:197–202.
Priestley M. B. Spectral Analysis and Time Series (1981) New York: Academic Press.
Sakai H., Sakaguchi F. Simultaneous confidence bands for the spectral estimate of two-channel autoregressive processes. J. Time Ser. Anal. (1990) 11:49–56.
Sakhanenko A. I. On the accuracy of normal approximation in the invariance principle. Siber. Adv. Math. (1991) 1:58–91.
Tomà
ek L. Asymptotic simultaneous confidence bands for autoregressive spectral density. J. Time Ser. Anal. (1987) 8:469–91.
CiteULike 
Connotea 
Del.icio.us What's this?
| ||||||||||||||||||||||||||||||||||||||||||||||||