ar-irls documentation

[DemianVera]

Is your feature request related to a problem? Please describe.

Hi all!

I have a small question about the documentation of the ar-irls function and the article cited there (Barker et al. 2013). The documentation of this function says: "The code should work well with 10-30 Hz data", but Barker's article shows the advantages of applying iteratively reweighted autoregressive least squares (AR-IRLS) even though the data used were sampled at 4 Hz.
The documentation may have been updated, but the paper cited remained the same.

No solution needed, just an answer.

Thanks!

P.S.: I think in the last update something was changed in the function change_stimulus_duration, because when applied to data from Dr. Hupperts tutorial, it throws some errors (line 45).

[alkvi]

I would also be curious if there's additional details about the AR-IRLS implementation regarding relationship to >=10Hz data, other than increased model order as mentioned in the article.