# Baseline correction in TF analysis

**URL:** <https://mne.discourse.group/t/baseline-correction-in-tf-analysis/2020>\
**Category:** Mailing List Archive (read-only)\
**Tags:** list-archive\
**Created:** [December 8, 2019, 1:21pm UTC](https://mne.discourse.group/t/baseline-correction-in-tf-analysis/2020 "2019-12-08T13:21:19Z")\
**Posts on this page:** 2\
**Page:** 1

<div class="post-metadata">

**Author:** ![system](https://global.discourse-cdn.com/free1/uploads/mne/original/1X/85cc6bd2b69cb698a166dc6d880fb550510d0144.jpeg) [@system](https://mne.discourse.group/u/system)\
**Post date:** [December 8, 2019, 1:21pm UTC](https://mne.discourse.group/t/baseline-correction-in-tf-analysis/2020/1 "2019-12-08T13:21:19Z")

</div>

External Email - Use Caution

Hi all,  
&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;I am a freshman in time-frequency analysis. I run a time-frequency  
analysis with tfr\_morlet. I found the results are large differences in  
doing baseline correction before or after average epochs. I was wondering  
what is the difference between these two steps?  
&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;Thank you very much

---

<div class="post-metadata">

**Author:** ![system](https://global.discourse-cdn.com/free1/uploads/mne/original/1X/85cc6bd2b69cb698a166dc6d880fb550510d0144.jpeg) [@system](https://mne.discourse.group/u/system)\
**Post date:** [December 8, 2019, 8:46pm UTC](https://mne.discourse.group/t/baseline-correction-in-tf-analysis/2020/2 "2019-12-08T20:46:14Z")

</div>

External Email - Use Caution

Hi Tao,

I am assuming you mean baseline correction on the output of the tfr\_morlet  
function? Since you are looking at power and it's not linear, the order of  
the operation should indeed matter. Regarding which is recommended, my  
intuition would be that averaging in the time-frequency domain is better  
when you are looking for subtle oscillations that are not necessarily  
completely aligned in phase. It's what we call induced power in MNE.

Mainak