Save % trimmed in CalculateImpliedTimescales

[mpharrigan]

Would it be useful when calculating implied timescales to also (maybe optionally) save some information on ergodic trimming, specifically the percent of data that is left for each lag time. It seems that this information is calculated anyways, and it would just be a matter of writing it to disk

[rmcgibbo]

This is a good idea. The question, IMO, is what format the output from CalculateImpliedTimescales should be in. The current two column format doesn't really let you add extra data...

If we changed the format to something like a csv file with

lagtime, ts1, ts2, ... tsn, fraction_trim, [other_data]\n
1, 10.0, 8.0, .... 3.0, 0.90, ...\n
...

maybe it would be more extensible, so we could add more fields.

[kyleabeauchamp]

I agree. My vote is to use pandas.read_csv to have our input files be self-documenting csv files.

[rmcgibbo]

We can do this without creating a pandas dependency.

[kyleabeauchamp]

Yes, we can. But this might be a situation where it make sense to use a DataFrame to keep track of data+metadata.