Statistical reduction of EEG data

Analyzing multi-channel EEG can be a daunting chore due to the volume of information that must be examined. For this reason, it is often more productive to reduce the data set by limiting the analysis to certain features or by selecting a subset of the data according to a certain criteria. One traditional method is to use frequency domain techniques to focus upon certain aspects of the data. However, when the frequency domain is used to analyze EEG data, the temporal domain features are lost. In this paper we examine an alternative method of data reduction and channel grouping using a statistical approach which retains time information.