Michael Platzer

please provide a reproducible example

Interesting! I haven't seen such cases myself. Could you share the plotted timing patterns of these customers? And what are your estimated t and gamma parameters? As I haven't run...

@rhshadrach sorry, my bad, I've overlooked that. A meaningful title has now been added.

I suspect the issue is related to this change (taken from the 2.2.0 release notes): > integer dtypes with missing values are cast to NumPy float dtypes and NaN is...

yes, you are right, it's still happening for main ``` >>> import pandas as pd >>> pd.__version__ '3.0.0.dev0+355.g92a52e231' >>> pd.Series([12, pd.NA], dtype="Int64[pyarrow]").astype("string[pyarrow]") 0 12.0 1 ```