[Eeglablist] Data Classification Accuracy vs p-value

[Scott Makeig]

The relationship depends on the measure used in the algorithm's objective
function... as well as on the pdf of measure differences.  The relationship
will typically be monotonic, but near certainly not linear.

Scott

On Thu, Apr 14, 2016 at 5:00 PM, Iman Mohammad-Rezazadeh <
irezazadeh at ucdavis.edu> wrote:

> Hi,
>
> It is not the pure EEG question:
>
> Let’s assume that we have two groups of subjects ‘A ‘and ‘B’ and for each
> group we have same set of EEG features [x1 x2 …xm]. If we run a
> multivariate test and then find  there is a significant statistical
> difference between A and B, can we conclude that if we apply a data
> classification method ( such as Support Vector Machine or such) on them
> then we will get high classification ratio?
>
> In other words, I wonder if there is any relationship between the p-values
> among two groups and the classification accuracy /performance .
>
> Thanks
>
> ============================================
>
> *Iman M.Rezazadeh, Ph.D. , M.Sc., B.Sc.*
>
> Senior Research Associate III
>
> UCLA David Geffen School of Medicine
>
> Semel Institute for Neuroscience and Human Behavior
>
> 760 Westwood Plaza, Ste 47-448
>
> Los Angeles, CA  90095
>
> http://www.linkedin.com/pub/iman-m-rezazadeh/10/859/840/
>
>
>
>
>
>
>
>
>
> Iman
>



-- 
Scott Makeig, Research Scientist and Director, Swartz Center for
Computational Neuroscience, Institute for Neural Computation, University of
California San Diego, La Jolla CA 92093-0961, http://sccn.ucsd.edu/~scott
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://sccn.ucsd.edu/pipermail/eeglablist/attachments/20160414/e3e242ea/attachment.html>