Adam Li

Re adding unit tests: - you can experiment w/ creating a dataset that should tests the functionality you have added. For example, https://github.com/mne-tools/mne-connectivity/blob/d71b0672c1573110ac747bb8b26469ea4857e4e8/mne_connectivity/vector_ar/tests/test_var.py#L28-L86 creates a noisy dataset, which is used...

> I've never worked with unit tests before. What would be a good unit test to add for the new features and how do I add that? As for the...

> I'm now having some trouble making the unit-tests. I noticed how you have a correct A matrix in your `create_noisy_data()` function which you can use to test whether the...

> Now the problem is that RidgeCV does not seem to outperform OLS in this sample dataset. In addition, many of the eigenvalues are dissimilar from the actual sample eigenvalues....

Hey @witherscp just checking in on this to see if you need any assistance? If there is trouble figuring out a rigorous scenario where this is tested, we can always...

Sure, a PR with a unit-test is welcome!

### TODO list for MNE-Connectivity 1. [ ] a PR to the docs page pointing to Frites 2. [ ] a PR to enable exporting of all Connectivity data structures...

### Container Discussion ^ tagging this as a container discussion. > What do you think? Thanks! Overall this is definitely in the right direction I think. See below for specific...

I think the `plot_circle` addition would be nice if you can add it. However, I would also add a stipulation that this really doesn't imply "connectivity" but is simply fitting...

Yep I would say doing the bootstrap is fine, but needs to come with a warning/caveat because it's a common mistake that ppl don't fully specify their assumptions when doing...