Detrending series

A time series can be detrended by using the Operations=>Detrend Data item in the series' plot window. Selecting this item creates a Detrend Time Series popup window.

This popup allows you to subtract the D.C. bias from data, which if not removed gives rise to significatn power at 0 Hz of a periodogram. Detrending should therefore be carried out before creating a periodogram of your time series.

There are two options: If the data shows no long term trend, it is best to simply subtract the mean and divide by the standard deviation. This gives a dataset witha mean of zero and a standard deviation of one. Otherwise, it is best to subtract a low-order polynomial fit to the data, since if the long term variation is not removed a Fourier Transform will inject significant power at the frequency of the variations.

Here (above) we show a data set that has been detrended using a low order polynomial.

It should be noted that if two nights data show significantly different D.C. levels you may also wish to consider detrending them seperately before combining them using the Operations=>Combine Series item.