SFB 303 Discussion Paper No. A - 121

Author: Franke, J., and W. Härdle
Title: On Bootstrapping Kernel Spectral Estimates
Abstract: We consider an application of bootstrap ideas to kernel estimates of spectral densities of stationary time series. We prove that the bootstrap principle holds in the same sense as in parametric and nonparametric estimation of regression curves. The general approach is illustrated with specific applications like a bootstrap method for choosing the bandwidth of a kernel estimate. In particular, we show that this method provides a consistent and asymptotically optimal bandwidth.
Keywords: bootstrap, spectral estimation, kernel spectrum estimates, bandwidth selection
Creation-Date: 1987
Unfortunately this paper is not available online. Please contact us to order a hardcopy.

SFB 303 Homepage

28.09.1999, Webmaster