ISSN:
1573-9686
Source:
Springer Online Journal Archives 1860-2000
Topics:
Medicine
,
Technology
Notes:
Abstract We developed a distribution-free, nonparametric statistical test (the permutation-rank test) for verifying the exiscence of significant peaks in power spectral ordinates calculated on noisy, short, time series. In this paper, we demonstrate the application of the test in the case of periodic series generated by the in numero perturbation of the amplitude of a cosine wave of known, single, constant frequency. The test was able to detect the underlying oscillation when it was amplitude-corrupted by noise generated by a standard uniform distribution, with a noise-to-signal ratio of 1 or less. It was shown that the permutation test generated a white process from the time series being tested. We concluded that any time series, if sampled for the periodicities of interest in a manner appropriate for the constraints imposed by the Nyquist theorem in the presence of unwanted random signal (noise), is suitable for standard time series analysis and application of this new permutation test for the significance of a spectral ordinate. The advantages of the new test over the standard methods are that it requires no assumptions about the data; does not require replicate analysis of a reference sample; takes into account both observation and system noise; and, at least in some instances, is more sensitive.
Type of Medium:
Electronic Resource
URL:
http://dx.doi.org/10.1007/BF02363068
Permalink