Open BenjaminBorn opened 7 years ago
I've been wondering for a while whether it wouldn't make more sense to move the (P)ACF functionality to a package like TimeSeries. It feels a little out of place here. But regardless of where it ends up, confidence intervals seem desirable.
I would tend to agree, except for the fact that the time series eco system is still in its infancy with many unmaintained packages. I therefore like to have the (P)ACF functionality in StatsBase for the moment where it is easily found by Julia novices. We can then move it out later.
Would there be interest in a PR that adds confidence intervals to the
autocor
andpacf
functions? Maybe with a keyword so that they are not computed by default.This would make it more convenient to create (partial) autocorrelogram plots.