What currency correlation means
Currency correlation describes how the prices of different currency pairs tend to move relative to each other. If both pairs usually rise and fall together, they are positively correlated. If they usually move in opposite directions, they are negatively correlated. If their moves show no steady link, the correlation is weak or close to zero.
Correlation is a statistical description of past behaviour. It does not tell you where any pair will go next, and it does not mean any pair causes the other to move.
If terms like base currency, quote currency or pip are new to you, see Currency pairs and the Trading glossary.
Why pairs sharing a currency move together or opposite
Every pair is a price of a base currency against a quote currency. When both pairs contain the same currency, news about that currency moves both of them.
Same currency, same side: positive correlation
EUR/USD and GBP/USD both have the US dollar as the quote currency. When the dollar weakens broadly, both pairs tend to rise. When it strengthens, both tend to fall. The euro and the pound are also tied by trade and geography, which adds to the link.
AUD/USD and NZD/USD behave similarly. Australia and New Zealand have closely linked economies, and both currencies often react in a similar way to commodity prices and global risk mood.
Same currency, opposite side: negative correlation
EUR/USD and USD/CHF both contain the dollar, but on opposite sides. In EUR/USD the dollar is the quote currency; in USD/CHF it is the base. A broad dollar rally pushes EUR/USD down and USD/CHF up. The result is a strong negative correlation over many periods.
The same logic applies to pairs like GBP/USD and USD/CAD, although the link is usually looser because the Canadian dollar also reacts to oil.
Summary table
| Pair combination | Shared currency | Position of shared currency | Usual direction of correlation |
|---|---|---|---|
| EUR/USD and GBP/USD | USD | Quote in both | Positive |
| AUD/USD and NZD/USD | USD | Quote in both | Positive |
| EUR/USD and USD/CHF | USD | Quote, then base | Negative |
| GBP/USD and USD/CAD | USD | Quote, then base | Negative, often weaker |
These are typical tendencies, not fixed rules. For more on how pairs are grouped, see Majors, minors, exotics.
How correlation stacks exposure without you noticing
Many traders open positions on several pairs and believe they have spread their risk. If the pairs are strongly correlated, the positions behave like a single large trade.
Example: a long position on EUR/USD and a long position on GBP/USD of similar size. Both are, in effect, bets against the US dollar. If the dollar strengthens, both positions lose at the same time. Your real exposure to the dollar is the sum of both positions, not the size of each.
The reverse case also matters. A long EUR/USD and a long USD/CHF partly cancel each other because of their negative correlation. You pay spread and possibly commission and swap on both, yet the net market exposure may be small.
With leverage, this hidden stacking can use up margin quickly. Leveraged forex and CFD trading can lose money fast, and correlated positions can make those losses arrive together. On Exness, the margin call level on the Standard account is 60% and on the Pro account it is 30%, while stop out on both is listed at 0%. These levels can change, and 0% stop out not supported in some countries; may vary, so check the current specification for your account in the Exness personal area or in the platform.
Practical checks before adding a position:
- List the currencies you are already long and short, not just the pairs.
- Add up exposure per currency across all open trades.
- Treat strongly correlated positions as a single trade when sizing risk.
Position sizing is covered in Risk management, and you can work out lot size and margin with the tools on Calculators.
Ready to try it on a real or demo account?
Partner link · trading carries a high risk of loss
Correlation changes over time
Correlation is not stable. A link that held for months can weaken or even reverse. Common reasons:
- Central bank policy. When interest rate paths of different countries diverge, their currencies can stop moving together. See Fundamental analysis.
- Commodity prices. Currencies of commodity exporters can follow raw material prices more closely in some periods than in others.
- Risk mood. In periods of market stress, many pairs suddenly move together as traders shift to perceived safe currencies.
- Local events. Elections, data surprises or interventions can break a pair away from its usual partners.
For this reason, a correlation value is only useful together with the period it was measured over. A short window reacts quickly but is noisy; a long window is smoother but slow to show change. Comparing both helps you see whether a relationship is stable or shifting.
How to compute correlation yourself
You do not need a live correlation matrix to understand the method. The standard measure is the Pearson correlation coefficient.
- Collect prices. Export closing prices for both pairs over the same dates and the same timeframe, for example daily closes from MetaTrader 5. See MetaTrader 5 for how the platform works.
- Convert to returns. Calculate the percentage change from each close to the next. Using returns rather than raw prices avoids false correlation caused by long trends.
- Calculate the coefficient. The coefficient is the covariance of both return series divided by the product of their standard deviations. Spreadsheet programs do this with a built-in correlation function such as CORREL.
- Read the result. The coefficient sits on a fixed scale. The top end means the pairs moved in perfect step, the bottom end means they moved in perfectly opposite directions, and the middle means no linear relationship.
- Repeat on a rolling window. Recalculate as new data arrives to see how the link changes.
You can practise this kind of analysis without real money on a Demo account.
Frequently asked questions
What is forex correlation in simple terms?
It is a measure of how closely the price moves of different currency pairs follow each other, either in the same direction or in opposite directions.
Are correlated pairs a trading signal?
No. Correlation describes past behaviour. It does not predict future prices and should not be read as a reason to open any position.
Does opening positions on several pairs reduce risk?
Only if the pairs are not strongly correlated. Positive correlation can stack exposure to a single currency; negative correlation can cancel positions out while you still pay trading costs.
Why did a correlation I relied on stop working?
Correlation shifts with interest rates, commodity prices, risk mood and local events. Always check the measurement period and recalculate regularly.
Where can I find Exness conditions that matter for correlated positions?
Margin call, stop out, leverage and costs depend on the account type and can change. Check the current specification in the Exness personal area or in the trading platform, and see Account types for an overview.
