[Eeglablist] ERD/ERS: log ratio with negative std_erspplot values

Carmen Dang c1dang at psych.ryerson.ca
Sun May 10 18:29:08 PDT 2020


Dear eeglablist,

I'd like to calculate the ERD/ERS of a frequency band with the following formula: log(A/R) where A is the power in the frequency band of interest in the period after the event and R is the power in the preceding baseline or reference period?. I'm using the values obtained from [STUDY erspdata] = std_erspplot(STUDY,ALLEEG,'channels', {'C4' 'Cz' 'C3'});, but some of these values are negative so I cannot take the logarithm of the ratio. Is there a transformation I should apply to the data (e.g. add a constant to all values)? Has 'erspdata' from std_erspplot been transformed similar to this post https://sccn.ucsd.edu/pipermail/eeglablist/2012/004755.html??

Thank you,
Carmen



More information about the eeglablist mailing list