Kevin Sheppard
Kevin Sheppard
Not sure if this would be a dream or a nightmare - anyway to add an option to allow select tags/structures to skipped from conversion in `conf.py`?
> I recognize I'm late to this, but out of curiosity, why use pyarrow string arrays instead of using numpy structured types for unicode strings (e.g., `dtype='
Seems pretty simple. Are the dta files produced by Stata or something else?
Just to check - are the dta files produced by Stata?
One more question: Does stata read all of the dta files correctly?
Do we have documentation that there are no differences between 110 and 111? This seems to be the assumption here.
Can you rebase and ping on green?
Thanks. LGTM.
Stata uses a wide variety of resolutions. This is the key function: https://github.com/pandas-dev/pandas/blob/ac5587cf092315d5cd30612e25c10c4c0325b107/pandas/io/stata.py#L240 There is a note that once other resolutions are implemented, other than ns, that it would make...
The mean is jointly estimated with the variance parameters. If you want the exact in-sample mean, you would need to first demean the data using the rolling mean, and then...