## Using currency correlations to your advantage

### A Wealth of Knowledge

Yonela Ngulugulu **23** Aug 2018

Currencies move to an extent in accord with other currencies

To be a good trader/investor a grasp of one's overall portfolio's sensitivity to market volatility It is important to understand you whole portfolio's sensitivity to market volatility to be a good trader/investor. This especially holds true when one is trading the Forex market. Why you ask? Well, because no single currency is traded alone, currencies are traded in pairs. A comprehension of these currency correlations and how they impact or affect other currencies, one can use such information to control their overall portfolio's exposure.

## Contents

## Defining correlation

It is easy to understand the reason for the correlation of currency pairs: Say for example one is trading the British Pound **(GBP)** against the Japanese Yen **(JPY)**. One is simply trading a derivative of the **GBP/USD** and **USD/JPY** pairs. Consequently, **GBP/JPY** should be correlated to either one of them if not both of them already. Although, it is important to note that the correlations between currencies stems from more than just the fact that they are traded in pairs. Although some currency pairs may move in tandem, however, other currency pairs may move in a contrary direction, which at the core is as the result of more complicated phenomenon.

Correlation refers to two variables that move in tandem to one another. It is a statistical assessment of the relation between two trading instruments. The correlation-coefficient ranges between -1 and +1. The word "correlation coefficient" refers to the statistic measuring how closely two variables co-vary. Variables can vary from -1 (perfect negative correlation) to 0 which means there is no correlation at all and lastly +1 (perfect positive correlation). So +1 refers to an instance whereby two securities will move in the same direction 100% of the time. Conversely, i correlation of -1 refers to an instance whereby two securities will move opposite to one another 100% of the time. Lastly o correlation of 0 refers to an instance whereby there is no correlation whatsoever between securities, therefore the move amid the securities is completely random.

## Reading the correlation table

Using the information mentioned above, now lets look at the following tables. Each of the tables below are showing correlations between the major currency pairs. This data is based on the recent market trading in the Forex market

EUR/USD | AUD/USD | USD/JPY | GBP/USD | NZD/USD | USD/CHF | USD/CAD |
---|---|---|---|---|---|---|

1 Month | 0.74 | 0.44 | 0.95 | 0.84 | -1.00 | -0.31 |

3 Months | 0.77 | -0.42 | 0.90 | 0.69 | -0.96 | -0.18 |

6 Months | 0.50 | -0.34 | 0.66 | 0.66 | -0.90 | -0.22 |

1 Year | 0.93 | -0.55 | 0.89 | 0.96 | -0.97 | -0.90 |

AUD/USD | EUR/USD | USD/JPY | GBP/USD | NZD/USD | USD/CHF | USD/CAD |
---|---|---|---|---|---|---|

1 Month | 0.74 | 0.71 | 0.80 | 0.93 | -0.72 | -0.80 |

3 Months | 0.77 | -0.17 | -0.57 | 0.92 | -0.84 | -0.57 |

6 Months | 0.50 | -0.59 | -0.82 | 0.88 | -0.74 | -0.82 |

1 Year | 0.93 | -0.58 | -0.97 | 0.99 | -0.97 | -0.97 |

The table above shows indicates that over 1 month the **EUR/USD** and the **GBP/USD** had a very effective correlation which was indicated at 0.95. This then means that when the **EUR/USD** rallies, the **GBP/USD** followed in that regard/accord, this had happened for about 95% of the time. Over the past six months the correlation was somewhat weaker than the month 1 (0.66). However, long term wise, measured over a year the correlation became stronger again.

On the other hand, we have the **EUR/USD** and **USD/CHF** in which there exist an almost perfect negative correlation of -1.00. This simply translates to the fact that about 100% of the time when the **EUR/USD** dropped the **USD/CHF** rallied. This correlation amid the two holds true in when measured over a sustained period, say a period of 1 year simply because the of the correlation figures remaining comparatively stable.

NZD/USD | EUR/USD | AUD/USD | USD/JPY | GBP/USD | USD/CHF | USD/CAD |
---|---|---|---|---|---|---|

1 Month | 0.84 | 0.93 | 0.76 | 0.86 | -0.81 | -0.72 |

3 Months | 0.69 | 0.92 | 0.09 | 0.71 | -0.77 | -0.63 |

6 Months | 0.66 | 0.88 | -0.42 | 0.22 | -0.84 | -0.73 |

1 Year | 0.96 | 0.99 | -0.57 | 0.86 | -0.97 | -0.96 |

USD/CHF | EUR/USD | AUD/USD | USD/JPY | GBP/USD | NZD/USD | USD/CAD |
---|---|---|---|---|---|---|

1 Month | -1.00 | -0.72 | -0.40 | -0.95 | -0.81 | 0.28 |

3 Months | -0.96 | -0.84 | 0.31 | -0.88 | -0.77 | 0.35 |

6 Months | -0.90 | -0.78 | 0.50 | -0.52 | -0.84 | 0.52 |

1 Year | -0.97 | -0.97 | 0.63 | -0.87 | -0.97 | 0.95 |

USD/CAD | EUR/USD | AUD/USD | USD/JPY | GBP/USD | NZD/USD | USD/CHF |
---|---|---|---|---|---|---|

1 Month | -0.31 | -0.80 | -0.70 | -0.35 | -0.72 | 0.28 |

3 Months | -0.18 | -0.57 | -0.44 | -0.11 | -0.63 | 0.35 |

6 Months | -0.22 | -0.82 | 0.36 | 0.11 | -0.73 | 0.52 |

1 Year | -0.90 | -0.97 | 0.58 | -0.88 | -0.96 | 0.95 |

Though, correlations do not remain fixed or solid. As an example, lets explore the relationship between **USD/CAD** and **USD/CHF**. With a coefficient of 0.95, these two currencies had a notable/significant positive correlation for the previous year however (2017), this has changed ever since and the change or is a significant one in the previous month (April 2018). The co-efficiency has dropped down to 0.28. This could be as a result of a number of different factors that could cause a significant reaction for certain currencies in the shorter term. An example would be say a rally in oil prices (This affects the Canadian and United States economies in particular), or the hawkishness of the **BoC**.

## Correlations change

It is obvious that correlations amid different currencies thus changes over time. This then makes for aligning ones trades with the shift in correlations crucial. Macroeconomic influences and general sentiment on certain country's economies tends to change too overtime and sometimes may change in about a day. This is to say that, a near-perfect correlation may not be as such long term wise. As such, it is for this reason that it is important to consider a 6 months correlation between currencies. This then allows for a more detailed out and better view and understanding of a relationship between currencies, as such, this tends to be more accurate. The change in the correlation of two currencies could be due to a number of reasons/factors, one of the most common or is a catalyst for such a change would be a difference between monetary policies of different countries. Another factor might be a currency's sensitivity to commodity prices furthermore, a country's unique economic and political factors.

## How to calculate correlations yourself

The most effective way to gauge or update yourself with regards to the strength of correlation between pairs is to calculate them yourself. Although this may sound somewhat difficult, on the contrary, it is much simpler than you think.

To do this, you will need a spreadsheet program, a good example of this is "Microsoft Excel". Numerous charting platforms allow for downloading past daily prices, which one can simply download and move to their Excel spreadsheet. Once transferred the information onto an Excel spreadsheet, use the correlation function which is as follows: CORREL(range 1 and 2) Much like the tables above, include the: 1 Year, 6 months, 3 months, 1 month readings as these give out the most informative perspective of the correlations and differences over time.

Here is the correlation-calculation process reviewed step by step:

- The first thing to do is get the pricing information for your two chosen currency pairs i.e
**EUR/USD**and**USD/CHF** - Thereafter, create two separate columns, each with one of the pairs written at the top. Furthermore, populate the columns with the previous daily prices measured over a time period one is analysing.
- Below one of the columns in an empty slot, write =CORREL(
- Highlight all of the data in one of the pricing columns; you should get a range of cells in the formula box.
- Type in comma
- Repeat steps 3-5 for the other currency
- Close the formula so that it looks like =CORREL(A1:A50,B1:B50)
- The number that is produced represents the correlation between the two currency pairs

Even though correlations change, it is not necessary to update your numbers every day; updating once every few weeks or at the very least once a month is generally a good idea.