Table of Contents
Pandas has a built-in method for calculating percentile ranks called .rank(). It takes a numerical column of data and returns the percentile rank of each value in the column, from 0 to 100. It is simple to use and can be applied to many different types of data. For example, you could use it to calculate the percentile rank of any given student’s test score. You can also use it to compare the percentile ranks of different groups of data, such as comparing the scores of two different classes.
The percentile rank of a value tells us the percentage of values in a dataset that rank equal to or below a given value.
You can use the following methods to calculate percentile rank in pandas:
Method 1: Calculate Percentile Rank for Column
df['percent_rank'] = df['some_column'].rank(pct=True)
Method 2: Calculate Percentile Rank by Group
df['percent_rank'] = df.groupby('group_var')['value_var'].transform('rank', pct=True)
The following examples show how to use each method in practice with the following pandas DataFrame:
import pandas as pd
#create DataFrame
df = pd.DataFrame({'team': ['A', 'A', 'A', 'A', 'A', 'A', 'A',
'B', 'B', 'B', 'B', 'B', 'B', 'B'],
'points': [2, 5, 5, 7, 9, 13, 15, 17, 22, 24, 30, 31, 38, 39]})
#view DataFrame
print(df)
team points
0 A 2
1 A 5
2 A 5
3 A 7
4 A 9
5 A 13
6 A 15
7 B 17
8 B 22
9 B 24
10 B 30
11 B 31
12 B 38
13 B 39
Example 1: Calculate Percentile Rank for Column
The following code shows how to calculate the percentile rank of each value in the points column:
#add new column that shows percentile rank of points
df['percent_rank'] = df['points'].rank(pct=True)
#view updated DataFrame
print(df)
team points percent_rank
0 A 2 0.071429
1 A 5 0.178571
2 A 5 0.178571
3 A 7 0.285714
4 A 9 0.357143
5 A 13 0.428571
6 A 15 0.500000
7 B 17 0.571429
8 B 22 0.642857
9 B 24 0.714286
10 B 30 0.785714
11 B 31 0.857143
12 B 38 0.928571
13 B 39 1.000000
Here’s how to interpret the values in the percent_rank column:
- 7.14% of the points values are equal to or less than 2.
- 17.86% of the points values are equal to or less than 5.
- 28.57% of the points values are equal to or less than 7.
And so on.
Example 2: Calculate Percentile Rank by Group
The following code shows how to calculate the percentile rank of each value in the points column, grouped by team:
#add new column that shows percentile rank of points, grouped by team
df['percent_rank'] = df.groupby('team')['points'].transform('rank', pct=True)
#view updated DataFrame
print(df)
team points percent_rank
0 A 2 0.142857
1 A 5 0.357143
2 A 5 0.357143
3 A 7 0.571429
4 A 9 0.714286
5 A 13 0.857143
6 A 15 1.000000
7 B 17 0.142857
8 B 22 0.285714
9 B 24 0.428571
10 B 30 0.571429
11 B 31 0.714286
12 B 38 0.857143
13 B 39 1.000000
- 14.3% of the points values for team A are equal to or less than 2.
- 35.7% of the points values for team A are equal to or less than 5.
- 57.1% of the points values for team A are equal to or less than 7.
And so on.