np.nan in Pandas: Why missing values break comparisons

Missing data in Python Pandas? We use nan (“not a number”), which comes from NumPy.

np.nan is a float, but not a normal one:

np.nan + 10           # nan
np.nan * 12345        # nan
sum([10, 20, np.nan]) # nan
np.nan == 0           # False
np.nan == np.nan      # False