Key points are not available for this paper at this time.
SUMMARY The problem of testing for a jump in the spectral function of a stationary time series is considered. The test based on the greatest periodogram ordinate divided by an estimate of the spectral density is discussed and a modification is suggested which consists essentially of taking out regression on the harmonic with frequency 2πj/n, before computing the estimate of the spectral density at that frequency. The regression can be approximately effected by using the corresponding periodogram ordinate.
E. J. Hannan (Sat,) studied this question.
Synapse has enriched 5 closely related papers on similar clinical questions. Consider them for comparative context: