Dealing with missing values

List-wise Deletion

Use to drop NA’s:

df.dropna(how="any") # to drop all rows with at least one missing value.
# You can pass axis=1 to drop column.
df.dropna(subset=["Col1"]) # to drop all rows with missing value on Col1

Use to fill NA’s:

df["Col"].fillna(
	value="None", inplace=True
)

Mask notnull by df["ValueGiven"] = df["Col1"].notnull().

Replacing

Numerical Values

One may use mean or median, but it can mess up the variance and covariance.