Analytics
Correlation Matrix
Measures how strongly each pair of the picked numeric columns moves together, from −1 to 1, as a matrix with one row per column.
When to use it
Use it to spot related measures before modelling (collinear features) or to find which drivers move with an outcome.
In a pipeline
File InputCorrelation MatrixExport Excel
Settings
| Setting | Type | Required | Default or hint |
|---|---|---|---|
| Columns | select | Yes | Pick two or more numeric columns |
| Method | select | — | pearson |
| Decimals | number | — | 3 |
Example
Spend, visits and returns
- Columns=Spend, Visits, Returns
- Decimals=2
| Spend | Visits | Returns |
|---|---|---|
| 10 | 2 | 5 |
| 20 | 4 | 3 |
| 30 | 5 | 4 |
| 40 | 8 | 1 |
| Column | Spend | Visits | Returns |
|---|---|---|---|
| Spend | 1 | 0.98 | -0.83 |
| Visits | 0.98 | 1 | -0.92 |
| Returns | -0.83 | -0.92 | 1 |
Worth knowing
- Pearson measures straight-line relationships; Spearman compares rank order, so it also catches curved but steady ones and resists outliers.
- Each pair uses the rows where both columns are numbers; a constant column gives a blank.
- Correlation is not causation — and values near ±1 between features warn that a regression will struggle to tell them apart.