search

Mensen / ept_TFCE-matlab

Advanced EEG Statistics
27 stars 9 forks source link

Simulated datasets and interaction between factors #15

Open Em-Fl opened 6 years ago

Em-Fl commented 6 years ago

Hey Mensen, first of all thank you for all your previous help, I come again with questions.

1) Is it possible to have access to the simulated datasets that you use in your paper of this tool? 2) I was wondering if it's possible to modify your scripts in order to define a specific statistic. For example to find clusters where there is an interaction between two factors, not any interaction but a specific one between the defined factors?

Thank you!

Mensen commented 6 years ago

Hi!

  1. I'll have to look through some old hard drives I have around, but if I can find them again you are more than welcome to have them.
  2. Its certainly possible to calculate another statistic... its not an easy option or anything... but there are only a few points in the script where the statistical value is calculated, once for the original ordering, and again for each of the permuted datasets. However, the second part of your question makes me think that perhaps you mean something different than just changing the test statistics (e.g. instead of T-values, using a Mann-Whitney U test)... Since you presumably just have two factors, there is only one interaction term between them... so there is no selecting between "specific ones". I suppose you mean testing specific conditions between each other after finding a significant interaction? If so, then you can just run a simpler ept_TFCE on the two conditions of particular interest to be compared just like a usual post-hoc comparisons test after an ANOVA.

Does any of that make sense?