MultiIndex (hierarchical indexing) in Data Analysis Python - Mini Project: Build & Apply

Working with MultiIndex (hierarchical indexing) in pandas

📖 Scenario: You work in a small store that sells fruits in different cities. You have sales data for each fruit in each city for two months. You want to organize this data so you can easily find sales by city and fruit.

🎯 Goal: You will create a pandas DataFrame with a MultiIndex using city and fruit. Then you will select sales data for a specific city and fruit using the MultiIndex.

📋 What You'll Learn

Create a pandas DataFrame with MultiIndex from city and fruit

Use a tuple list to create the MultiIndex

Select sales data for a specific city and fruit using .loc

Print the selected sales data

💡 Why This Matters

🌍 Real World

Stores and businesses often have sales data organized by multiple categories like location and product. MultiIndex helps organize and access this data easily.

💼 Career

Data analysts and scientists use MultiIndex in pandas to handle complex datasets with multiple levels of grouping, making data analysis more efficient.

1

Create the sales data dictionary

Create a dictionary called sales_data with these exact entries: ('New York', 'Apple'): [100, 120], ('New York', 'Banana'): [90, 110], ('Los Angeles', 'Apple'): [80, 95], ('Los Angeles', 'Banana'): [70, 85]. The lists represent sales for two months.

Data Analysis Python

# Create the sales_data dictionary with city and fruit as keys and sales lists as values
# Your code here

Hint

Use tuples as keys in the dictionary to represent city and fruit pairs.

2

Create the MultiIndex and DataFrame

Create a variable called index using pd.MultiIndex.from_tuples() with the keys of sales_data. Then create a DataFrame called df with the values of sales_data, the index as index, and columns named ['Month 1', 'Month 2'].

Data Analysis Python

sales_data = {
    ('New York', 'Apple'): [100, 120],
    ('New York', 'Banana'): [90, 110],
    ('Los Angeles', 'Apple'): [80, 95],
    ('Los Angeles', 'Banana'): [70, 85]
}
import pandas as pd

# Create the MultiIndex from the keys of sales_data
# Create the DataFrame df with sales_data values, index, and columns ['Month 1', 'Month 2']
# Your code here

Hint

Use list(sales_data.values()) to get the data for the DataFrame.

3

Select sales data for New York Apples

Use df.loc with the tuple ('New York', 'Apple') to select the sales data for apples in New York. Store this in a variable called ny_apple_sales.

Data Analysis Python

sales_data = {
    ('New York', 'Apple'): [100, 120],
    ('New York', 'Banana'): [90, 110],
    ('Los Angeles', 'Apple'): [80, 95],
    ('Los Angeles', 'Banana'): [70, 85]
}
import pandas as pd

index = pd.MultiIndex.from_tuples(sales_data.keys(), names=['City', 'Fruit'])
df = pd.DataFrame(list(sales_data.values()), index=index, columns=['Month 1', 'Month 2'])

# Select sales data for New York apples using df.loc
# Your code here

Hint

Use df.loc[('New York', 'Apple')] to get the row for New York apples.

4

Print the selected sales data

Print the variable ny_apple_sales to display the sales for apples in New York.

Data Analysis Python

sales_data = {
    ('New York', 'Apple'): [100, 120],
    ('New York', 'Banana'): [90, 110],
    ('Los Angeles', 'Apple'): [80, 95],
    ('Los Angeles', 'Banana'): [70, 85]
}
import pandas as pd

index = pd.MultiIndex.from_tuples(sales_data.keys(), names=['City', 'Fruit'])
df = pd.DataFrame(list(sales_data.values()), index=index, columns=['Month 1', 'Month 2'])

ny_apple_sales = df.loc[('New York', 'Apple')]

# Print the ny_apple_sales variable
# Your code here

Hint

Use print(ny_apple_sales) to show the sales data.