How to Calculate Rolling Sum in pandas: Simple Guide

To calculate a rolling sum in pandas, use the rolling(window) method on a DataFrame or Series followed by sum(). This computes the sum of values over a sliding window of specified size across your data.

Syntax

The basic syntax to calculate a rolling sum is:

• data.rolling(window).sum()

Here, data is your pandas Series or DataFrame column.

window is the size of the moving window (number of rows) to sum over.

The method returns a new Series or DataFrame with the rolling sums.

python

rolling_sum = data.rolling(window=3).sum()

Example

This example shows how to calculate a rolling sum with a window size of 3 on a pandas Series.

python

import pandas as pd

data = pd.Series([1, 2, 3, 4, 5, 6])
rolling_sum = data.rolling(window=3).sum()
print(rolling_sum)

Output

0    NaN
1    NaN
2    6.0
3    9.0
4   12.0
5   15.0
dtype: float64

Common Pitfalls

Common mistakes when calculating rolling sums include:

• Not setting the window size correctly, which changes the result.
• Expecting results for the first few rows where the window is not full; these return NaN by default.
• Using rolling on DataFrames without specifying the column, which can cause confusion.

To avoid NaN at the start, you can use min_periods=1 to get sums with fewer values.

python

import pandas as pd

data = pd.Series([1, 2, 3, 4, 5])

# Wrong: no min_periods, first two are NaN
print(data.rolling(window=3).sum())

# Right: min_periods=1 to avoid NaN
print(data.rolling(window=3, min_periods=1).sum())

Output

0    NaN
1    NaN
2    6.0
3    9.0
4   12.0
dtype: float64
0     1.0
1     3.0
2     6.0
3     9.0
4    12.0
dtype: float64

Quick Reference

Summary tips for rolling sums in pandas:

• Use rolling(window).sum() to get rolling sums.
• window defines how many rows to include.
• Use min_periods=1 to avoid NaN at the start.
• Works on Series and DataFrame columns.

Key Takeaways

Use pandas' rolling(window).sum() to calculate rolling sums easily.

The window size controls how many values are summed at each step.

By default, rolling sums return NaN until the window is full; use min_periods=1 to change this.

Rolling sums work on both Series and DataFrame columns.

Always check your window size and min_periods to get expected results.