Data & CSV
Grouped CSV rolling window
Add trailing moving summaries while keeping groups and input order intact.
A clear answer. Here’s how.
Uses the latest N records encountered within each exact group, including the current record and initial partial windows. Computes mean, sum, minimum or maximum; does not sort dates or infer a time interval. Requires numeric non-blank cells. Rows × window is capped at 2 million operations; ordinary CSV limits also apply.
See a worked example
Worked example
A fixed demonstration, separate from the values you enter above.
- Your CSV
ID,Name,Team,Amount,Date 001,Ada,Design,120,2026-01-05 002,Sam,Writing,80,2026-01-06 003,Jo,Design,150,2026-01-07
- Column separator
- Comma
- Value column
- Amount
- Group column
- Team
- Window size
- 3
- Summary
- mean
- New column heading
- Rolling value
Example result
| ID | Name | Team | Amount | Date | Rolling value |
|---|---|---|---|---|---|
| 001 | Ada | Design | 120 | 2026-01-05 | 120 |
| 002 | Sam | Writing | 80 | 2026-01-06 | 80 |
| 003 | Jo | Design | 150 | 2026-01-07 | 135 |
Trailing windows follow input order separately within each exact group. Initial partial windows are included; blank numeric values are errors.