A MATLAB system to load EEG data, add a dummy signal, and then try to classify with a variety of feature selection methods. It is intended to work with BCI competition data (http://www.bbci.de/competition/iv/). Feature selection methods include PCA, ADEN, ADENZ, GADEN, and GADENZ.
If you use this code, please cite: LaRocco, J., Innes, C. R., Bones, P. J., Weddell, S., & Jones, R. D. (2014, August). Optimal EEG feature selection from average distance between events and non-events. In 2014 36th Annual International Conference of the IEEE Engineering in Medicine and Biology Society (pp. 2641-2644). IEEE.