
The cov() method in pandas computes the pairwise covariance of columns, excluding NA/null values. The returned data frame is the covariance matrix of the columns of the DataFrame. Both NA and null values are automatically excluded from the calculation.
Example –
Let’s read a dataset to work with.
import pandas as pd
from sklearn import datasets
housing = datasets.fetch_california_housing()
df = pd.DataFrame(housing.data, columns=housing.feature_names)
df.head()

Now to create a covariance matrix in pandas, we can use the df.cov() method.
cov_matrix = df.cov()
cov_matrix
