You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
If a dataset has a longer time series, the default number of possible values for tau to test may not be enough. I don't know what the threshold ratio of timesteps to taus should be. This should probably be figured out analytically. E.g. reduce # of meshpoints relative to timesteps until estimated tau is significantly different from estimate with very high number of meshpoints?
The text was updated successfully, but these errors were encountered:
Suggested here: ropenscilabs/statistical-software-review#2 (comment)
If a dataset has a longer time series, the default number of possible values for tau to test may not be enough. I don't know what the threshold ratio of timesteps to taus should be. This should probably be figured out analytically. E.g. reduce # of meshpoints relative to timesteps until estimated tau is significantly different from estimate with very high number of meshpoints?
The text was updated successfully, but these errors were encountered: