Merging on index in Pandas

Introduction

Merging on index helps you combine two tables by matching their row labels instead of columns. It is useful when the row labels carry important information.

You have two tables with related data and their row labels (indexes) match.

You want to join data based on unique IDs stored as row indexes.

You want to combine time series data where dates are the index.

You want to add columns from one table to another using the index as the key.

Syntax

Pandas

pd.merge(left, right, left_index=True, right_index=True, how='inner')

Set left_index=True and right_index=True to merge on the indexes of both tables.

The 'how' parameter controls the type of join: 'inner', 'left', 'right', or 'outer'.

Examples

Merge two dataframes on their indexes using an inner join by default.

Pandas

pd.merge(df1, df2, left_index=True, right_index=True)

Merge two dataframes on their indexes including all rows from both (outer join).

Pandas

pd.merge(df1, df2, left_index=True, right_index=True, how='outer')

Merge two dataframes on their indexes keeping all rows from the left dataframe.

Pandas

pd.merge(df1, df2, left_index=True, right_index=True, how='left')

Sample Program

This example merges two dataframes on their indexes. Only rows with matching indexes 'b' and 'c' appear in the result.

Pandas

import pandas as pd

# Create first dataframe with index
df1 = pd.DataFrame({'A': [1, 2, 3]}, index=['a', 'b', 'c'])

# Create second dataframe with index
df2 = pd.DataFrame({'B': [4, 5, 6]}, index=['b', 'c', 'd'])

# Merge on index with inner join
result = pd.merge(df1, df2, left_index=True, right_index=True, how='inner')

print(result)

Important Notes

Merging on index is useful when the index uniquely identifies rows.

If indexes do not match, the result depends on the 'how' parameter.

You can reset the index to merge on columns instead if needed.

Summary

Merging on index joins tables by their row labels.

Use left_index=True and right_index=True to merge on indexes.

The 'how' parameter controls which rows appear in the result.

Practice

1. What does setting left_index=True and right_index=True do when merging two pandas DataFrames?

easy

A. It merges the DataFrames by ignoring the indexes completely.

B. It merges the DataFrames using their column names as the key.

C. It merges the DataFrames by concatenating columns without any key.

D. It merges the DataFrames using their row indexes as the key.

Solution

1. Step 1: Understand the role of indexes in merge

   Indexes are row labels in pandas DataFrames. Using left_index=True and right_index=True tells pandas to join on these row labels instead of columns.

2. Step 2: Effect of these parameters in merge

   When both are True, pandas matches rows based on their index values, combining rows with the same index from both DataFrames.

3. Final Answer:

   It merges the DataFrames using their row indexes as the key. -> Option D

4. Quick Check:

   Merging on indexes = left_index=True and right_index=True [OK]

Hint: Indexes join rows by labels, not columns [OK]

Common Mistakes:

• Thinking merge always uses columns as keys
• Confusing index merge with concatenation
• Ignoring the need to set both left_index and right_index

2. Which of the following is the correct syntax to merge two DataFrames df1 and df2 on their indexes with an inner join?

easy

A. pd.merge(df1, df2, on='index', how='inner')

B. pd.merge(df1, df2, left_index=True, right_index=True, how='inner')

C. pd.merge(df1, df2, index=True, how='inner')

D. pd.merge(df1, df2, left_on='index', right_on='index', how='inner')

Solution

1. Step 1: Identify correct parameters for index merge

   To merge on indexes, pandas requires left_index=True and right_index=True. Using on='index' or left_on='index' is incorrect because 'index' is not a column name.

2. Step 2: Confirm the join type

   The how='inner' parameter specifies an inner join, which keeps only matching index rows.

3. Final Answer:

   pd.merge(df1, df2, left_index=True, right_index=True, how='inner') -> Option B

4. Quick Check:

   Use left_index and right_index for index merge [OK]

Hint: Use left_index and right_index, not 'on' for index merge [OK]

Common Mistakes:

• Using on='index' instead of left_index=True
• Confusing index merge with column merge
• Forgetting to set both left_index and right_index

3. Given the following DataFrames:

df1:
       A
idx
1      10
2      20
3      30


df2:
       B
idx
2     200
3     300
4     400

What is the result of pd.merge(df1, df2, left_index=True, right_index=True, how='inner')?

medium

A. {'idx': [2, 3], 'A': [20, 30], 'B': [200, 300]}

B. {'idx': [1, 2, 3], 'A': [10, 20, 30], 'B': [null, 200, 300]}

C. {'idx': [1, 2, 3, 4], 'A': [10, 20, 30, null], 'B': [null, 200, 300, 400]}

D. {'idx': [2, 3, 4], 'A': [20, 30, null], 'B': [200, 300, 400]}

Solution

1. Step 1: Understand inner join on indexes

   An inner join keeps only rows where the index exists in both DataFrames. Here, indexes 2 and 3 are common.

2. Step 2: Combine matching rows

   Rows with index 2 have A=20 and B=200, index 3 have A=30 and B=300. Rows with index 1 and 4 are excluded.

3. Final Answer:

   {'idx': [2, 3], 'A': [20, 30], 'B': [200, 300]} -> Option A

4. Quick Check:

   Inner join keeps only common indexes [OK]

Hint: Inner join keeps only matching index rows [OK]

Common Mistakes:

• Including rows with indexes not in both DataFrames
• Confusing inner join with outer join
• Expecting nulls in inner join result

4. Consider this code snippet:

import pandas as pd

df1 = pd.DataFrame({'A': [1, 2]}, index=[0, 1])
df2 = pd.DataFrame({'B': [3, 4]}, index=[1, 2])

result = pd.merge(df1, df2, left_index=True, right_index=True, how='outer')
print(result)

What is the error in this code?

medium

A. No error; the code runs and merges correctly.

B. Missing 'on' parameter causes a KeyError.

C. left_index and right_index cannot be used together.

D. DataFrames must have the same index values for merge.

Solution

1. Step 1: Check parameters used in merge

   The code uses left_index=True and right_index=True correctly to merge on indexes with how='outer'.

2. Step 2: Understand outer join behavior

   Outer join includes all indexes from both DataFrames, filling missing values with NaN. This is valid and runs without error.

3. Final Answer:

   No error; the code runs and merges correctly. -> Option A

4. Quick Check:

   left_index and right_index merge works fine [OK]

Hint: Outer join merges all indexes, no error with index merge [OK]

Common Mistakes:

• Assuming missing 'on' causes error when using index merge
• Thinking left_index and right_index cannot be combined
• Believing indexes must match exactly for merge

5. You have two DataFrames:

df1:
       Score
Student
Alice     85
Bob       90
Charlie   78


df2:
       Grade
Student
Bob       B
Charlie   C
David     A

You want to merge them so that all students from df1 appear, and matching grades from df2 are added. Missing grades should show as NaN. Which code achieves this?

hard

A. pd.merge(df1, df2, left_index=True, right_index=True, how='right')

B. pd.merge(df1, df2, left_index=True, right_index=True, how='inner')

C. pd.merge(df1, df2, left_index=True, right_index=True, how='left')

D. pd.merge(df1, df2, left_on='Student', right_on='Student', how='outer')

Solution

1. Step 1: Identify merge type for all df1 rows

   To keep all rows from df1 and add matching data from df2, use a left join.

2. Step 2: Merge on indexes

   Both DataFrames use 'Student' as index, so set left_index=True and right_index=True to merge on indexes.

3. Final Answer:

   pd.merge(df1, df2, left_index=True, right_index=True, how='left') -> Option C

4. Quick Check:

   Left join on indexes keeps all df1 rows [OK]

Hint: Left join keeps all left DataFrame rows [OK]

Common Mistakes:

• Using inner join which drops unmatched df1 rows
• Using right join which keeps df2 rows instead
• Merging on columns instead of indexes