A correlation matrix shows how every pair of variables moves together. Enter two or more series to calculate Pearson or Spearman coefficients, test their significance, and rank the strongest relationships.
How it works
Measure the strength and direction of linear (Pearson) or rank (Spearman) relationships between two or more series, with significance tests for every pair.
Pearson r = Σ(x − x̄)(y − ȳ) ÷ √[Σ(x − x̄)² · Σ(y − ȳ)²]
Spearman ρ = Pearson r computed on the ranks (ties receive average ranks)
t = r·√((n − 2) ÷ (1 − r²)), two-tailed p-value with n − 2 degrees of freedomCommon uses
- Selecting drivers for forecasting and regression models
- Portfolio and risk analysis
- Exploring relationships in operational data
Worked example
The calculator above opens with this example. Change any input to see the results update instantly, or select Reset to return to these figures.
| Series | Revenue | 412 398 436 445 468 452 441 459 487 503 498 541 Marketing spend | 38 35 42 44 47 45 41 44 50 53 51 58 Headcount | 120 121 121 123 124 126 126 127 129 130 132 133 Interest rate | 5.25 5.25 5.30 5.20 5.10 5.05 5.00 4.90 4.85 4.80 4.75 4.70 |
| Method | Pearson (linear) |
| Strongest relationship | Revenue & Marketing spend |
| Series | 4 |
| Observations per series | 12 |
| Significant pairs (p < 0.05) | 6 |
Correlation does not imply causation; trending series often correlate simply because both grow over time. Test changes (first differences) to check.
Interpreting the results
- Use Spearman when relationships are monotonic but not linear, or when outliers distort Pearson.
- With small samples even large coefficients may not be significant; check the p-value.
- Highly correlated drivers in the same regression create multicollinearity; keep one or combine them.
Frequently asked questions
What is the difference between Pearson and Spearman correlation?
Pearson correlation measures the strength of a linear relationship and is sensitive to outliers. Spearman correlation measures any consistently increasing or decreasing relationship using ranks, which makes it more robust to outliers and curved patterns.
What counts as a strong correlation?
As a common rule of thumb, an absolute coefficient above 0.7 is strong, 0.4 to 0.7 is moderate, and 0.2 to 0.4 is weak. Context matters: in some fields a correlation of 0.3 is meaningful.
Why might two unrelated series be highly correlated?
Series that both trend over time, such as revenue and headcount at a growing company, often correlate simply because both rise with time. Correlating period-to-period changes instead of levels helps reveal genuine relationships.
Related calculators
- Linear Regression & Trend Forecast: Slope, intercept, R², p-values and forecasts with confidence and prediction intervals.
- Descriptive Statistics Calculator: Mean, median, standard deviation, quartiles, skewness and kurtosis.
- Forecast Accuracy Calculator: MAPE, WMAPE, MAD, RMSE, bias and tracking signal from actuals and forecasts.
- Hypothesis Test Calculator (t, z, Chi-Square): One- and two-sample t-tests, proportion z-tests and chi-square tests of independence.
- Monte Carlo Profit Simulator: Simulate thousands of profit outcomes with P10/P50/P90 and probability of loss.