permutation stat for one subejct

[Karlo Gonzales]

Hello all,

i was wondering if I could have your advice on the statistical test on 60 channels of EEG for one subject:

I have 6 minutes rest-EEG data from one subject in pre post design, and want to check if the power spectrum has changed between pre and post.

i am considering these steps

1- after pre-processing EEG data, epoching continues data in a 2-second segments 

1a,  compute power, run t-test

2- Bootstrapping steps

 2a- permuting segments between pre and post 

 2b- compute power, run t-test

 3b-repeat steps 2a&2b for 1000 times

3-  compare p-value from step1a with p-values distribution from step 2b

Thank you in advance.

Reza

[Mike X Cohen]

Hi Reza. I think your option 2 looks good; that way you don't need to worry about assumptions. 

(Minor point: the procedure you describe is called permutation testing, not bootstrapping; those are related but different statistical approaches.)

--
You received this message because you are subscribed to the Google Groups "AnalyzingNeuralTimeSeriesData" group.
To unsubscribe from this group and stop receiving emails from it, send an email to analyzingneuraltimes...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/analyzingneuraltimeseriesdata/e04717cb-0745-4b78-b249-37e712de9de5n%40googlegroups.com.

--

Mike X Cohen, PhD
Company: sincxpress.com
Personal: mikexcohen.com

LinkedIn: linkedin.com/in/mikexcohen