A currency correlation is the degree by which one currency is interrelated with another. The currency correlation is represented on a numeric scale ranging from -1 to +1, in the same fashion as the correlation coefficient. The numeric values included in a currency correlation represent the degree of correlation. Each value is generated via the mathematics related to the computing of a statistical "correlation coefficient."
On the surface, the term "currency correlation" can appear complex and a bit intimidating. However, it's actually fairly straightforward aside from the mathematics used behind the scenes to create the currency correlation value itself.
In order to understand clearly what a currency correlation is, we first must understand what a correlation is, and more specifically, what a "correlation coefficient" is. The term "correlation coefficient" is defined as being a statistical measure of the interdependence between two random variables. Correlation coefficients are represented numerically on a scale between -1 and +1; -1 represents an inverse relationship between the variables, while +1 represents a positive linear correlation. In the case of currency and forex trading, the random variables involved in the correlation are the individual market values of the currency pairs.
It is important to note that in the forex market, currencies are not traded in isolation; they are traded in pairs. Each pair's valuation is based on the constantly changing exchange rates between the two currencies included in the pair. It is a common practice for forex traders, or investors, to be actively trading multiple currency pairs simultaneously. If this is the trading strategy, then the trader or investor must be aware of the correlations concerning each currency pair that is being traded. Without a functioning currency correlation defining the degree of the relationship between traded pairs, it is very likely that a trader could unintentionally increase risk when attempting to diversify the portfolio.
One of the most important aspects to remember regarding currency correlations is that they are constantly changing in value. As exchange rates of the currency pairs themselves fluctuate, the relationships between the pairs evolve. The correlation of a currency pair today is not necessarily the correlation of the currency pair tomorrow.
To account for the constant fluctuations of currency correlations, calculations using different time frames, or periods are performed. A correlation for a currency pair can be calculated using short-term intraday time frames, while also being calculated on a daily, weekly, monthly or yearly basis. The distinction in time frame is important as to the function of the correlation within a given strategy. For example, a short-term intraday trader would be more interested in a currency correlation value that is calculated on a minute-by-minute basis, while a long-term investor would be more interested in longer time frame correlations.
Scale Of Measurement
As mentioned earlier, currency correlations are expressed using a scale with a range between -1 and +1. The numbers represent the varying degree and specific type of correlation. There are three basic types of correlations: positive, negative and neutral.
A positive correlation expresses a linear relationship between two variables. For instance, if currency pair A moves up or down in value and currency pair B moves in the same direction as A, then the correlation is said to be positive. A correlation of +1 means that when A experiences a move, B also moves in the same direction 100% of the time. Usually, a perfect correlation of +1 is not present, thus a positive correlation of varying degree is represented by a value that lies between 0 and +1.
A negative correlation expresses an inverse relationship between two variables. In this case, when currency pair A experiences a price movement, currency pair B moves in the opposite direction. The value of -1 is used to represent the perfect inverse relationship. Again, most currency correlations are not 100% perfect, thus a negative correlation is represented with a value ranging from -1 to 0.
Lastly, a correlation of zero expresses a neutral relationship between two variables. So, when currency pair A experiences a change in pricing, currency pair B does not behave in any specific manner. A perfect zero correlation represents no correlation between A and B whatsoever, or a random relationship. Values that are close to zero are considered to be very slightly correlated; thus the closer to zero we move on the scale, the slighter the correlation.
Currency Correlation Tables
Now that we have defined what a currency correlation is, and how it is expressed, we can talk about where the data is located. Currency correlation data is often compiled for quick reference using data tables, heat maps, scatter-point diagrams, bar graphs and spreadsheets.
The currency correlation table is a useful and easily interpreted representation of correlation data. Basically, a currency correlation table is a data table that includes colour-coded correlation values for a specified period. Most commonly, a black colour represents a positive correlation and a red colour represents a negative correlation. The use of contrasting colours within the table format provides a quick and easy way to interpret a large amount of data as it pertains to the correlation of currencies.
One of the oldest fundamental correlations concerning forex currency pairs is the "regional correlation." The premise behind the "regional correlation" is the geographic proximity certain currency pairs have in respect to one another. Basically, the closer that two currency pairs are geographically, the more likely a positive correlation between the two exists.
For instance, EUR/USD is a heavily traded major pair on the forex exchange, and it's a good indicator of the euro's strength in comparison to the United States dollar. A less popular regional pair would be the Swiss franc to the United States dollar, CHF/USD. During the period of January 1999 to August 2001, EUR/USD and CHF/USD had an exceptional daily positive correlation of .95. Essentially, CHF/USD and the EUR/USD moved in lockstep every day during this period. The +.95 value, coupled with the extensive period of time in which the correlation existed, is a good illustration of an extremely high positive correlation. If recognised in a timely manner, this correlation would have been very useful in the development of trading and investment strategies concerning EUR/USD and CHF/USD currency pairings. Past performance is no indication of future results.
The search for measurable correlations and the strategies in which they are incorporated knows no bounds. Whether attempting to find a short-term arbitrage opportunity, or diversifying a large investment portfolio, currency correlations play an integral role in a trader or investor's strategy. At the end of the day, the currency correlation can be a valuable tool in deciding which currency pairs complement one another, and which pairs are best used to hedge risk.
FXCM Research Team
FXCM Research Team consists of a number of FXCM's Market and Product Specialists.
Articles published by FXCM Research Team generally have numerous contributors and aim to provide general Educational and Informative content on Market News and Products.
Retrieved 13 Dec 2015 https://www.thefreedictionary.com/correlation+coefficient
Retrieved 13 Dec 2015 https://arxiv.org/pdf/cond-mat/0303306v1.pdf