QUB | Archaeology and Palaeoecology | The 14Chrono Centre

Manual for psimpoll and pscomb

Menu

Menu M: data analyses

Menu Mf: Fourier
Fourier analysis produces this display:
4 Fourier analysis (off)                         [off]

Enter '0' for on, or '1' for off
Fourier analysis converts the data from the time domain to the frequency domain. This form of time series analysis is normally only available for samples that are evenly spaced in time, which is far from true for most pollen data. However, a technique devised by Lomb to generate a periodogram works with unevenly sampled data. I have implemented this, on an experimental basis, from Press et al. (1992). Results are written to the data analysis file, giving a string of values for the importance of each taxon at a range of frequencies. The numbers can be plotted by `cutting' the results out of that file and creating a new file that can be presented to psimpoll. The number of taxa depends on how many you selected when the program ran. The number of `samples' (now `frequencies') is 5 × the number of samples in the original dataset. The 2-character code for input Item 4 is `lf', resulting in appropriate labelling (see Data preparation).

This option may run slowly on older computers, perhaps 10 minutes per taxon for a dataset with 80 samples on a 386 processor. The length of time increases as n² where n is the number of samples.

If 0 is selected, you will see the following submenu:

1 Data transformation                            [0]
9 Leave this menu
Q Return to main menu
Me.1 allows a data transformation before Fourier analysis is carried out. The submenu is the same as that described for zonation (menu Mc.5), and discussed for principal components analysis (menu Md.3).

Return to main menu.


Back to contents page
Copyright © 1995-2007 K.D. Bennett

Archaeology and Palaeoecology | 42 Fitzwilliam St | Belfast BT9 6AX | Northern Ireland | tel +44 28 90 97 5136


Archaeology and Palaeoecology | The 14Chrono Centre | URL http://www.qub.ac.uk/arcpal/ | WebMaster