Percentiles with np.percentile() in NumPy - Step-by-Step Execution

Concept Flow - Percentiles with np.percentile()

Start with data array

↓

Choose percentile value p

↓

Sort data in ascending order

↓

Calculate rank index for p

↓

Interpolate if needed

↓

Return value at percentile p

The process takes a data array and a percentile value, sorts the data, finds the position for that percentile, interpolates if needed, and returns the percentile value.

Execution Sample

NumPy

import numpy as np

data = np.array([10, 20, 30, 40, 50])
result = np.percentile(data, 40)
print(result)

Calculate the 40th percentile of a simple numeric array.

Execution Table

Step	Action	Data State	Calculation	Result
1	Start with data array	[10, 20, 30, 40, 50]	N/A	N/A
2	Choose percentile p	p = 40	N/A	N/A
3	Sort data	[10, 20, 30, 40, 50]	Already sorted	N/A
4	Calculate rank index	n=5	rank = (p/100)*(n-1) = 0.4*4 = 1.6	rank=1.6
5	Interpolate between index 1 and 2	values at 1=20, 2=30	20 + 0.6*(30-20) = 20 + 6 = 26	26.0
6	Return percentile value	N/A	N/A	26.0

💡 Percentile value 26.0 returned after interpolation at rank 1.6

Variable Tracker

Variable	Start	After Step 4	After Step 5	Final
data	[10, 20, 30, 40, 50]	[10, 20, 30, 40, 50]	[10, 20, 30, 40, 50]	[10, 20, 30, 40, 50]
p	N/A	40	40	40
n	N/A	5	5	5
rank	N/A	1.6	1.6	1.6
result	N/A	N/A	26.0	26.0

Key Moments - 2 Insights

Why do we interpolate between two data points for percentile 40?

Because the rank 1.6 is not an integer index, np.percentile interpolates between the values at index 1 and 2 (20 and 30) to find the exact percentile value, as shown in execution_table step 5.

Why is the data sorted before calculating the percentile?

Percentiles depend on the order of data. Sorting ensures we can find the correct position for the percentile, as shown in execution_table step 3.

Visual Quiz - 3 Questions

Look at the execution table, what is the rank calculated for the 40th percentile?

A 1.6

B 2.0

C 0.4

D 4.0

💡 Hint

Check the 'Calculate rank index' row in the execution_table.

At which step does interpolation happen in the execution table?

A Step 3

B Step 5

C Step 4

D Step 6

💡 Hint

Look for the step mentioning 'Interpolate' in the Action column.

If the percentile p was 50, what would be the rank index?

A 3.0

B 2.5

C 2.0

D 1.5

💡 Hint

Use the formula rank = (p/100)*(n-1) with n=5 and p=50.

Concept Snapshot

np.percentile(data, p) computes the p-th percentile of data.
Data is sorted first.
Rank index = (p/100)*(n-1).
If rank is not integer, interpolate between neighbors.
Returns the value at percentile p.

Full Transcript

This visual execution shows how np.percentile calculates the 40th percentile of a data array. First, the data is sorted. Then the rank index is calculated using the formula (p/100)*(n-1), where n is the number of data points. For p=40 and n=5, rank is 1.6. Since 1.6 is not an integer, interpolation happens between the values at index 1 and 2, which are 20 and 30. The interpolated value is 26.0, which is returned as the 40th percentile. Variables like data, p, n, rank, and result change as the steps progress. Key moments include understanding why sorting and interpolation are needed. The quizzes test understanding of rank calculation, interpolation step, and rank for a different percentile.