Bootstrap confidence bands for spectra and cross-spectra

A nonparametric method for setting confidence intervals and confidence bands for spectra and cross-spectra of stationary weakly dependent time series is presented. The proposed methodology involves using a bootstrap resampling scheme that was recently developed for use in time series problems. A computing algorithm is provided, along with guidelines on its practical application. Finally, results of simulations with artificially generated data are shown as an illustration of the method's potential.<
>