Selecting data with MultiIndex in Pandas - Practice Problems & Coding Challenges

Challenge - 5 Problems

❓ Predict Output

intermediate

Selecting rows by level value in MultiIndex

What is the output of this code snippet that selects rows where the first level of the MultiIndex is 'B'?

Pandas

import pandas as pd

index = pd.MultiIndex.from_tuples([('A', 1), ('A', 2), ('B', 1), ('B', 2)])
data = pd.DataFrame({'value': [10, 20, 30, 40]}, index=index)

result = data.loc['B']
print(result)

A

   value
0     30
1     40

B

       value
B 1     30
  2     40

C KeyError: 'B'

D

   value
1     30
2     40

💡 Hint

Use .loc with the first level key to select all rows under that level.

Using .loc['B'] on a MultiIndex DataFrame selects all rows where the first level index is 'B'. The resulting DataFrame has the second level index as its index.

❓ data_output

intermediate

Selecting data with .xs() method

Given the MultiIndex DataFrame below, what is the output of selecting data with .xs(2, level=1)?

Pandas

import pandas as pd

index = pd.MultiIndex.from_tuples([('A', 1), ('A', 2), ('B', 1), ('B', 2)])
data = pd.DataFrame({'value': [10, 20, 30, 40]}, index=index)

result = data.xs(2, level=1)
print(result)

A KeyError: 2

B

   value
A     20
B     40

C

   value
A 2   20
B 2   40

D

   value
1     20
2     40

💡 Hint

The .xs() method selects data at a particular level value across the other level.

Using .xs(2, level=1) selects all rows where the second level index is 2, returning a DataFrame indexed by the first level.

🔧 Debug

advanced

Why does this MultiIndex selection raise an error?

What error does this code raise and why?

Pandas

import pandas as pd

index = pd.MultiIndex.from_tuples([('X', 'a'), ('X', 'b'), ('Y', 'a'), ('Y', 'b')])
data = pd.DataFrame({'val': [1, 2, 3, 4]}, index=index)

result = data.loc[('X',)]
print(result)

A Returns rows with first level 'X' and all second level values

B TypeError: unhashable type: 'tuple'

C KeyError: ('X',)

D IndexError: too many indices

💡 Hint

Check how tuples are used to select MultiIndex rows with .loc.

Using .loc with a tuple containing only one element ('X',) looks for an exact match of that tuple in the index, which does not exist, causing a KeyError.

🚀 Application

advanced

Selecting multiple levels with .loc using slice

Given this MultiIndex DataFrame, which option correctly selects all rows where the first level is 'A' and the second level is between 2 and 3 inclusive?

Pandas

import pandas as pd

index = pd.MultiIndex.from_tuples([('A', 1), ('A', 2), ('A', 3), ('B', 2), ('B', 3)])
data = pd.DataFrame({'val': [10, 20, 30, 40, 50]}, index=index)

# Select rows here

A data.loc[('A', slice(2, 3))]

B data.loc[('A', [2, 3])]

C data.loc['A', 2:3]

D data.loc[('A', 2:3)]

💡 Hint

Use slice objects inside .loc for range selection on MultiIndex levels.

Using data.loc[('A', slice(2, 3))] correctly selects rows with first level 'A' and second level between 2 and 3 inclusive.

🧠 Conceptual

expert

Understanding MultiIndex partial indexing behavior

Which statement best describes the behavior of partial indexing with .loc on a MultiIndex DataFrame?

A Partial indexing with .loc selects all rows matching the provided prefix of index levels.

B Partial indexing with .loc requires specifying all levels of the MultiIndex exactly.

C Partial indexing with .loc always returns a Series, never a DataFrame.

D Partial indexing with .loc raises an error if fewer levels than the MultiIndex are specified.

💡 Hint

Think about how .loc behaves when given fewer keys than the number of index levels.

Partial indexing allows selecting all rows that match the given prefix of the MultiIndex levels, returning all matching rows.