Currency Correlation Calculator

How it works

What correlation measures

Pearson correlation measures how two series co-move around their own averages. A positive value means the two return series tended to move in the same direction during the sample. A negative value means they tended to move in opposite directions. It does not say one caused the other.

Use returns, not prices

For currency pairs, correlation must be calculated on percentage returns or another return definition, not on raw price levels. Raw price levels can trend together for reasons that disappear once the series is converted into period-to-period changes.

The Pearson formula

The calculator uses:

r = sum((x_i - mean x)(y_i - mean y)) / sqrt(sum((x_i - mean x)^2) * sum((y_i - mean y)^2))

When price inputs are selected, each price series is first converted to simple returns with price_t / price_{t-1} - 1.

Correlation percent and labels

The displayed correlation percent is round(r * 100). The label bands are: above +70 strong positive, +30 to +70 moderate positive, -30 to +30 weak, -70 to -30 moderate negative, and below -70 strong negative.

Worked example

For returns x = [0.01, 0.02, -0.01, 0.03, 0.00] and y = [0.02, 0.01, -0.02, 0.04, 0.01], Pearson correlation is about 0.87519. The correlation percent is 88, which falls in the strong positive band.

Important limitations

Correlation is sample-period sensitive. A daily 90-day window and a one-hour intraday window can give very different readings. Correlations can flip during stress regimes, and common USD exposure can create co-movement that is not a stable relationship.