Adaptive smoothing of the log-spectrum with multiple tapering

A hybrid estimator of the log-spectral density of a stationary time series is proposed. First, a multiple taper estimated is performed, followed by kernel smoothing the log-multiple taper estimate. This procedure reduces the expected mean square error by (/spl pi//sup 2//4)/sup 4/5/ over simply smoothing the log tapered periodogram. A data-adaptive implementation of a variable-bandwidth kernel smoother is given.