Kevin Sheppard

Seeing same in statsmodels on Windows. https://dev.azure.com/statsmodels/statsmodels-testing/_build/results?buildId=4410&view=logs&j=9c2ba6b6-9ef1-5300-06cd-4db9cc34ebf3&t=f07685ef-4dee-5864-e861-bb3b9ab4fc6b

@Carreau I get the same error when I `pip install ipython

I can also get this error when I reun specific tests: ``` ValueError: Plugin already registered: c:\git\statsmodels\statsmodels\conftest.py= ```

IMO you either go with `int64` or just make them `float`, if you want to move away from the idea of using the smallest int dtype that can represent the...

> The intended behavior can be achieved by specifying any signed dtype as you can see here: This isn't quite right. Any int dtype can wrap under the right conditions....

Another funny behavior of `get_dummies` ``` In [26]: pd.get_dummies(c,dummy_na=True) Out[26]: a b NaN 0 1 0 0 1 0 1 0 2 1 0 0 3 0 0 1 In...

I'm not sure this is a practical test in practice since as far as I can tell it is only designed for ARCH-type models and otherwise it requires model scores.

These models have not been implemented and so a volatility class would be needed for both.

If someone wanted to implement this I'd be happy to include it. I don't think I'm likely to implement it.

There are no immediate plans to add it. I worked on it a little in a branch but there isn't a full model. There was a lot of design thought...