df1: 
        Column 1  Column 2  Column 3  Column 4
Row 1       1.0       2.0       NaN       4.0
Row 2       5.0       6.0       7.0       8.0
Row 3       9.0       NaN      11.0      12.0
Row 4       NaN      14.0      15.0      16.0
Row 5       NaN      18.0      19.0       NaN
df1.isnull: 
        Column 1  Column 2  Column 3  Column 4
Row 1     False     False      True     False
Row 2     False     False     False     False
Row 3     False      True     False     False
Row 4      True     False     False     False
Row 5      True     False     False      True

As we can see the non-null values are indicated as False and the null values are indicated using True in the output of the DataFrame.isnull() function.

At this point, if we compute DataFrame.isnull().mean(), the True values in the DataFrame will contribute 1 to the sum and the False values will contribute 0. As a result, the mean value will indicate: (the total number of True values)/(the total number of True + False values). If we multiply this value with 100, we will get the percentage of null values in each column of the DataFrame.

import pandas
import numpy

list1 = [[1, 2, numpy.nan, 4], [5, 6, 7, 8], [9, numpy.nan, 11, 12], [numpy.nan, 14, 15, 16], [numpy.nan, 18, 19, numpy.nan]]
df1 = pandas.DataFrame(list1, index=["Row 1", "Row 2", "Row 3", "Row 4", "Row 5"], columns=["Column 1", "Column 2", "Column 3", "Column 4"])
print("df1: \n", df1)

print("df1.isnull: \n", df1.isnull())

print("Percentage of missing values: \n", df1.isnull().mean()*100)

The output of the above program will be like the following:

df1: 
        Column 1  Column 2  Column 3  Column 4
Row 1       1.0       2.0       NaN       4.0
Row 2       5.0       6.0       7.0       8.0
Row 3       9.0       NaN      11.0      12.0
Row 4       NaN      14.0      15.0      16.0
Row 5       NaN      18.0      19.0       NaN

df1.isnull: 
        Column 1  Column 2  Column 3  Column 4
Row 1     False     False      True     False
Row 2     False     False     False     False
Row 3     False      True     False     False
Row 4      True     False     False     False
Row 5      True     False     False      True

Percentage of missing values: 
Column 1    40.0
Column 2    20.0
Column 3    20.0
Column 4    20.0
dtype: float64

Hence, columns 1, 2, 3, and 4 contain 40%, 20%, 20%, and 20% missing values, respectively.

Now, let’s compute the percentage of missing values in the titanic dataset. We can use the following Python code to compute the …