Outlier detection with IQR in Pandas - Mini Project: Build & Apply

Outlier detection with IQR

📖 Scenario: You work as a data analyst for a retail company. You have sales data for different stores. Sometimes, some sales numbers are unusually high or low. These are called outliers. Detecting outliers helps you understand if there are errors or special cases in the data.

🎯 Goal: You will learn how to detect outliers in sales data using the Interquartile Range (IQR) method with pandas. You will create the sales data, calculate IQR, find outliers, and display them.

📋 What You'll Learn

Use pandas to create and analyze data

Calculate quartiles and IQR

Detect outliers using IQR method

Print the detected outliers

💡 Why This Matters

🌍 Real World

Outlier detection helps businesses find unusual data points that may indicate errors, fraud, or special events.

💼 Career

Data analysts and data scientists often use IQR to clean data and improve the quality of their analysis.

1

Create sales data with pandas

Create a pandas DataFrame called sales_data with one column named sales and these exact values: 100, 150, 200, 250, 300, 350, 400, 1000.

Pandas

# Create the sales_data DataFrame with the given sales values
# Your code here

Need a hint?

Use pd.DataFrame with a dictionary where the key is "sales" and the value is the list of numbers.

2

Calculate Q1, Q3 and IQR

Create three variables: Q1, Q3, and IQR. Calculate Q1 as the 25th percentile of sales_data["sales"], Q3 as the 75th percentile, and IQR as Q3 - Q1.

Pandas

import pandas as pd

sales_data = pd.DataFrame({"sales": [100, 150, 200, 250, 300, 350, 400, 1000]})

# Calculate Q1, Q3 and IQR below
# Your code here

Need a hint?

Use the quantile() method on the sales column to get Q1 and Q3.

3

Detect outliers using IQR

Create a variable called outliers that contains the rows from sales_data where the sales value is less than Q1 - 1.5 * IQR or greater than Q3 + 1.5 * IQR.

Pandas

import pandas as pd

sales_data = pd.DataFrame({"sales": [100, 150, 200, 250, 300, 350, 400, 1000]})

Q1 = sales_data["sales"].quantile(0.25)
Q3 = sales_data["sales"].quantile(0.75)
IQR = Q3 - Q1

# Create outliers DataFrame using IQR method
# Your code here

Need a hint?

Use boolean conditions combined with | inside the DataFrame filter brackets.

4

Print the detected outliers

Write a print statement to display the outliers DataFrame.

Pandas

import pandas as pd

sales_data = pd.DataFrame({"sales": [100, 150, 200, 250, 300, 350, 400, 1000]})

Q1 = sales_data["sales"].quantile(0.25)
Q3 = sales_data["sales"].quantile(0.75)
IQR = Q3 - Q1

outliers = sales_data[(sales_data["sales"]  (Q3 + 1.5 * IQR))]

# Print the outliers DataFrame
# Your code here

Need a hint?

Use print(outliers) to show the rows with outliers.