[Eeglablist] toolboxes for multiple testing of time-frequency data
arno at ucsd.edu
Wed Jan 15 18:13:41 PST 2014
thanks for your message. I would like to add to the list the excellent and widely used cluster method developed by Maris and Oostenveld (available in EEGLAB and Fieldtrip).
Maris, E., & Oostenveld, R. (2007). Nonparametric statistical testing of EEG- and MEG-data. Journal of Neuroscience Methods, 164, 177– 190. doi: 10.1016/j.jneumeth.2007.03.024.
According to Guillaume Rousselet and Cyril Pernet, this is the only method that adequately control for type I error rate when correcting for multiples comparisons (so if you have an uncorrected threshold p-value at 0.05, you corrected p-value threshold will be exactly 0.05 corrected for multiple comparisons - which might not be the case when using other methods). I tend to agree with them.
On Dec 21, 2013, at 1:22 AM, Archana Singh <sine.arc at gmail.com> wrote:
> Dear All,
> From time to time, I see people asking for options to correct for multiple comparison of statistics from high-dimensional time/frequency data, such as ERSP. A part of my research concerns the very issue of controlling false positives in EEG functional connectivity studies (see the references below). Detecting significance from a huge data in 3D (sensors, time, and frequency intervals) is like searching for a needle in the hay. Most multiple testing methods, including FDR, tend to control the false positives at the cost of hiding true positives - so we often end up loosing out on significance when it should in fact be evident.
> I would like to share some toolboxes that offer alternative options to maximizes the power without compromising on control of false positives. These options, called hierarchical FDR control and Optimal Discovery Procedure, are adopted from the approaches developed by Yekutieli and Storey, respectively, for their application to EEG data. See the references below for details.
> These toolboxes are in matlab and R, if you like to try them, send me an email.
> 1) Yekutieli, D., 2008. Hierarchical false discovery rate controlling methodology. Journal of the American Statistical Association 103 (481), 309–316.
> 2) Storey, J.D., 2007. The optimal discovery procedure: a new approach to simultaneous significance testing. Journal of the Royal Statistical Society: Series B (Statistical Methodology) 69 (3), 347–368.
> 3) Phillips, S., Takeda, Y., & Singh, A. K. Visual feature integration indicated by phase-locked frontal-parietal EEG signals. PLoS ONE 7(3): e32502, 2012
> 4) Singh A. K., Asoh H., and Phillips S., Optimal detection of functional connectivity from high-dimensional EEG synchrony data, NeuroImage, 58(1), 148-156, 2011.
> 5) Singh A. K. and Phillips S., Hierarchical control of false discovery rate for phase locking measures of EEG synchrony, NeuroImage, 50(1), 40-47, 2010
> Archana K Singh,
> ERATO Researcher,
> Functional and Chemosensory brain imaging group
> Department of Applied Biological Chemistry
> Graduate School of Agricultural and Life Sciences
> The University of Tokyo, Tokyo 113-8657
> (tel)+81-3-5841-5590, (fax)+81-3-5841-8024
> (email) archana at mail.ecc.u-tokyo.ac.jp
> Eeglablist page: http://sccn.ucsd.edu/eeglab/eeglabmail.html
> To unsubscribe, send an empty email to eeglablist-unsubscribe at sccn.ucsd.edu
> For digest mode, send an email with the subject "set digest mime" to eeglablist-request at sccn.ucsd.edu
-------------- next part --------------
An HTML attachment was scrubbed...
More information about the eeglablist