Bivariate Fourier (Cross Spectrum) Analysis Results - Advanced Tab
Select the Advanced tab of the Bivariate Fourier (Cross Spectrum) Analysis Results dialog box to access the options described here.
| Element Name | Description |
|---|---|
| Summary | Click this button to produce a summary spreadsheet containing the frequencies, periods, cosine and sine coefficients, periodogram values, spectral density estimates (computed according to the options selected in the Data windows for density estimates group box on the Advanced tab), cross-periodogram values (real and imaginary), cospectral density and quadrature spectrum values (computed according to the options selected in the Data windows for density estimates group box on the Advanced tab), the cross amplitude values, coherency values, gain values, phase spectrum values, and the weights used to produce the spectral density estimates. If the Highlight values larger than... check box is selected on the Advanced tab, then all values in the periodogram and spectral density columns that are larger than the specified value will be highlighted. For a description of how the values in this spreadsheet are computed, see the Summary description in the Bivariate Fourier (Cross Spectrum) Analysis Results dialog box topic. |
| Highlight values larger than | Select this check box to highlight values in the spreadsheet that are larger than that specified in the adjacent edit box. |
| Display N largest values | Click this button to display a Spreadsheet with the same statistics as those computed by the Summary option; however, the results will be shown only for the N (as specified in the edit box directly under this button) largest real cross-periodogram or imaginary cross-periodogram values, depending on the selection of the option buttons below the N= edit box. |
| Examine subset of periodogram | When reviewing the results for long time series, it is often useful to examine particular frequency ranges. This can be accomplished by selecting this check box, and then specifying the range of observations to review in the edit boxes below it (From, To). Note that the numbers specified here pertain to the numbers of rows in the Summary spreadsheet, so for example, if you specify 100 and 200, then rows 100 to 200 of the spreadsheet will be displayed (or plotted). In addition to the Summary spreadsheet option, the range selected here will also apply to all plots on this dialog. |
| Data windows for density estimates | As described in greater detail in the
Single Spectrum (Fourier) Analysis Overview, the periodogram (and cross-periodogram) values themselves are subject to substantial random fluctuation. A clearer picture of underlying periodicities often only emerges when examining the spectral and cospectral densities, that is, the frequency regions, consisting of many adjacent frequencies that contribute most to the overall periodic behavior of the series. The spectral and cospectral density estimates are computed by smoothing the periodogram values with a weighted moving average. Suppose the moving average window is of width m (which must be an odd number); the smoothers listed below are those that are most commonly cited in the literature (note: p = (m-1)/2). For more information on each window listed below, refer.
Daniell Tukey Hamming Parzen Bartlett If the User-defined option button is selected, you can enter the weights used for smoothing directly (click the button adjacent to User-defined to display a dialog to specify the weights). To assign a common value to all weights, enter it in the respective edit box, click the Apply button, then click OK to accept the weights. Note that the weights will always be standardized so that they sum to 1. For example, if you entered 1 2 3 2 1 as the user-defined weights, they will be converted to: 1/9 2/9 3/9 2/9 1/9 (so that they add up to 1). In many cases, all of the standard data windows will produce very similar results, but it is often useful to experiment with different windows of different widths, and to examine resulting periodogram and density plots. |