diff() for differences in Pandas

Introduction

The diff() function helps you find how much values change between rows or columns. It shows the difference step-by-step, making it easy to see trends or changes.

To see daily changes in stock prices from a list of prices.

To find how much sales increased or decreased each month.

To calculate the change in temperature from one day to the next.

To analyze the difference in scores between rounds in a game.

To detect sudden jumps or drops in sensor data over time.

Syntax

Pandas

DataFrame.diff(periods=1, axis=0)

periods sets how many steps to look back for difference (default is 1).

axis=0 means difference between rows; axis=1 means difference between columns.

Examples

Finds difference between each row and the previous row (default 1 period, axis=0).

Pandas

df.diff()

Finds difference between each row and the row 2 steps before.

Pandas

df.diff(periods=2)

Finds difference between each column and the previous column for each row.

Pandas

df.diff(axis=1)

Sample Program

This code creates a table of sales over 5 days. Then it adds a new column showing how much sales changed from the previous day.

Pandas

import pandas as pd

data = {'Day': [1, 2, 3, 4, 5],
        'Sales': [100, 120, 115, 130, 125]}
df = pd.DataFrame(data)

# Calculate daily sales difference
df['Sales_diff'] = df['Sales'].diff()

print(df)

Important Notes

The first difference value is always NaN because there is no previous row to subtract from.

You can use diff() on both Series and DataFrames.

Negative values mean a decrease, positive values mean an increase.

Summary

diff() shows how values change step-by-step.

Use it to find differences between rows or columns easily.

It helps spot trends, jumps, or drops in your data.