Missing values in df resulting in TypeError

[labanyamukhopadhyay]

Describe the problem

There is an issue with missing values in a dataframe that results in 'TypeError: boolean value of NA is ambiguous'. This may be due to modin not filling missing values before hand.

Source code

import statsmodels.api as sm

X_sm = sm.add_constant(X_drop_first)
model = sm.OLS(y.astype(float),X_sm.astype(int))
results = model.fit()
results.summary()

After adding this line X_drop_first.fillna(0, inplace=True) before the import statement, the following error occurs:
ValueError: unrecognized data structures: <class 'modin.pandas.series.Series'> / <class 'numpy.ndarray'>

[devin-petersohn]

Thanks @labanyamukhopadhyay for the report!

What is X_drop_first in this case?

[labanyamukhopadhyay]

X_drop_first is a modin dataframe with no clear missing values

[vnlitvinov]

@labanyamukhopadhyay would you be able to construct a self-contained reproducer (with input data)?