**How to Use Forex Correlation Stats**

On the Daily Forex Stats Page you will find forex correlation studies and volatility studies. Forex correlation stats may seem daunting, but a basic understanding of correlations can go a long way toward helping you to become a better trader. To learn how to use *volatility* data, please see: How to Use Forex Volatility Stats.

In fact, not understanding forex correlations when trading can be disastrous, but by understanding a bit more about correlations many pitfalls can be avoided…and possibly some additional strategies can be added to your forex trading arsenal. Correlations are used to help us determine which trades to take and which to filter out, as well as help us control risk.

**What is a Forex Correlation?**

A correlation is a measure of how much one currency moves with another. Correlations will run between -100 and +100, the former meaning they move in opposite directions to one another, and the latter meaning they move in the same direction.

The number between -100 and +100 shows the strength of the relationship. -100 shows two different pairs always moved inversely to each other over the period being tested. On the Daily Forex Stats page you can test correlations over varying lengths of time, and different time frames such as hourly, daily or weekly data. A reading of +86 shows there is a very strong correlation between two currency pairs–they move in the same direction very often, but not all the time.

A +100 correlation means two pairs move in the same direction. A -100 correlation means the pairs move in opposite directions. A correlation of 0 (zero) or a small positive or negative number (such as -30 or +25) means the pairs have no real correlation and if they do move together it is more likely to be random than anything significant. Typically a correlation of -/+ 70 is significant and noteworthy, while -/+ 80 is a strong correlation (or strong inverse correlation if negative).

Correlations do not measure magnitude. For example, the USDJPY and GBPJPY have a +98.7 correlation based on the chart below. They usually move in a similar direction, but that does not mean they move the same amount. Based on volatility data (a separate stat shown on the Daily Forex Stats page) the USDJPY currently moves 72.7 pips while the GBPJPY moves 137.4 pips per day. This is important when hedging and attempting to control risk, discussed later.

*Correlations are constantly changing, *usually at a slow pace. The exact numbers used in all these examples are subject to change on a daily basis. Always check for the newest correlation and/or volatility data. That said, certain pairs generally exhibit strong positive and inverse correlations to each even though the exact amount of correlation fluctuates over time. The statistics on the Daily Forex Stats are updated daily to reflect the forex correlations between pairs. The correlations are presented in a matrix as shown in the table below, and are presented for hourly, daily and weekly data.

*Daily Forex Correlation Table – December 30, 2013*

If we look down the EUR/USD column we see the EUR/USD has a strong inverse correlation of -95.9 to the USD/CHF. As one pair goes up the other will go down much of the time. The EUR/USD and GBP/USD usually share a significant or strong positive correlation, but at the time of this snapshot the correlation was only +58.2, which isn’t very significant. The weekly data on the Daily Forex Stats page though indicates that on the weekly time frame the correlation is +90, so over the longer-term the pairs still often move together.

Based on the chart there is no correlation between the EUR/USD and NZD/USD: +5.

The AUDUSD and USDJPY currently share a very strong inverse correlation (one moves up the other moves down) at -96.9.

Below are some guidelines and uses for forex correlation data.

**How to Use Forex Correlation Data **

–Short-term traders will find the hourly and daily correlation studies most useful. Long-term traders will want to monitor daily and weekly correlations.

–If you have multiple positions that are highly correlated (positive value over 70) it means that pairs move somewhat in tandem. This means you may be overexposed to one currency, even though the risk on each position is managed. Using the table above, if you are long the GBPUSD and the GBPJPY, these two pairs share a +91.9 correlation. That means they often move in the same direction, which means you may be potentially over-exposed to the GBP, because if the one of these pairs drop the other is likely to as well, resulting in two losing trades.

–You can hedge a trade in one currency pair with a trade in another that has a high (above 80) negative number. For example, you can take a long in the EURUSD and a long in the USDCHF; since they move in opposite direction one position hedges the other. Be aware of which currency is listed first though. The EURUSD and the USDCHF move in opposite directions but it is because the USD is only the base currency in one of the pairs. If you are long the EURUSD and short the USDCHF it is essentially means you have two short positions in the USD–double exposure. This can be good if it moves in your favor, bad if the USD moves against you.

–Just because two pairs are highly negatively or positively correlated does not mean they will completely offset each others losses when you are trying to hedge. Since each pair may move a different amount (more or less volatile), volatility is another factor which must be considered when looking at hedging.

–If you believe one currency will be strong, for instance the USD, you can buy the USD versus a number of other currencies that are somewhat or highly correlated. This will increase USD exposure, but each pair can provide different target, stops and risk. If you don’t like the trade set-up in one pair, you may be able to find a better setup in a different pair that is highly correlated. You get a similar trade but with a better set-up.

–When two or more pairs are highly correlated the correlation can be used as a confirmation for the moves which are occurring. For instance, if the GBPUSD makes a big move higher but the EURJPY doesn’t (+91.3 correlation at time of writing), it could warn that the move higher in the GBPUSD will soon fail. Keep in mind though the GBPUSD pair will have its own trend that affects, and is affected by, the EURJPY, and vice versa.

–Divergences from correlations are potential trading opportunities as well as potential warning signals that something significant is going on in one of the currencies. Pairs that diverge from long-term correlations may revert back and provide a trading opportunity, or it may signify a breakdown of the correlation.

I wish to expand on these topics by writing more in-depth articles on the topics discussed in this article. Please leave your questions or comments as this helps me determine where more information is needed.

**Forex Correlation Summary**

This is a brief and incomplete introduction to correlations. It is recommended that traders educate themselves further on correlations and acquaint themselves with statistics.

Refer to the Daily Forex Stats page to be aware of current correlations. See how positively and negatively correlated pairs interact with each other. You may find that being aware of correlations can help you control risk, find alternative trading strategies and alert you to potential dangers or opportunities.

**Download the Forex Strategies Guide for Day and Swing Traders 2.0 . Over 300 pages of Forex basics and 20+ Forex strategies for profiting in the 24-hours-a-day Forex market. This isn’t just an eBook, it’s a course to build your skill step by step.**

Would you suggest trading a basket of uncorrelated pairs or picking one correlated pair over the other for risk management purposes

I actually do both of these things. Typically I try to trade uncorrelated pairs, but I will occasionally take up to 2 trades that are highly correlated (but no more than 2) if I really like the setup. If I have three pairs with a valid trade setup, and they are all highly correlated, then I will trade the two with the best setup (based on risk:reward). This happens from time to time, but quite often I will just trades in pairs (with valid setups) in pairs that aren’t correlated.

A bit of flexibility is required though, because sometimes a lot of the pairs move together, so that is why I trade up to 2 highly correlated pairs and don’t worry about it.