Posts

Benefits of Portfolio Diversification

How to Measure Correlation Between Assets

In the previous episode, we showed you two ways to measure correlation. These were; coefficient of correlation (r) and coefficient of determination (R²). These are two widely used techniques for measuring correlation.

However, they do have their limitations.

When the quote currency in two fx pairs is the same, there will always be a correlation. However, due to the fundamental differences in base currency, they will also display uncorrelated behaviours.

This gives us an incorrect view whereby the (R²) number indicated a high level of correlation but the outlying data points highlighted the uncorrelated nature too. Talk about confusing.

A small change can have a big effect.

Today we’re going to make a subtle change that can provide a more robust view of the correlation between assets.

Instead of using price, we’re going to use the price difference. As an example, if we subtract an assets close price from the previous days’ close price, we can see the price change. It is this change in price that we will use for our new calculation.

Already you should be able to see the benefit of using this way. It allows us to compare price movements between the two assets.

We’ve given you a few examples of how correlation differs between assets and a few examples of different ways to measure correlation.

You can take this one step further by creating a correlation matrix in Excel for an indefinite number of assets.

Try doing the above to measure correlation between all the assets you have in your portfolio and see how diversified your asset selection is.

Darwinex helps you measure correlation of Darwins.

Darwinex has already done the hard work for you on the Darwinex platform by calculating the correlation between Darwins.

Remember a Darwin is an asset that tracks the performance of a traders underlying trading account, in real-time. Darwinex then manages the risk of the Darwin independently from the trader to target a max VaR (95%) of 6.5%.

To view the correlation between Darwins go to the correlation tab. It will show the correlation between the asset you have selected and a range of other Darwins. To check a particular darwin enter its ticker into the search box under the table to see the correlation between them.

Brought to you by Darwinex: UK FCA Regulated Broker, Asset Manager & Trader Exchange where Traders can legally attract Investor Capital and charge Performance Fees.

Risk disclosure:
https://www.darwinex.com/legal/risk-disclaimer


Content Disclaimer: The contents of this video (and all other videos by the presenter) are for educational purposes only, and are not to be construed as financial and/or investment advice.

Measure Correlation Portfolio Diversification Trading Strategy

Measuring Portfolio Diversification Correlation – A Common Mistake

Take care to avoid common mistakes when Measuring Portfolio Diversification Correlation. It is especially important to not view the results with blinkers on and ignore important information that is right in front of us.

Ultimately we want to improve the robustness of our trading strategies and Measuring Portfolio Diversification Correlation is a beneficial way to do this. But it is possible to wrongly make assumptions ignoring data, often unintentionally.

The easiest way to illustrate this is in Excel. Remember in the previous video we promised you, you don’t need to have experience in coding.

Well, we keep our promises.

You’ll need to first import the data of the assets you want to compare. Next up, pick the two assets you want to measure the correlation of. Input your formulas into excel and let it figure out the rest.

Next, display your results in a nice visually easy manner and you’re almost done. A couple of clicks later and you have an easy to see chart and the (R²) number.

Using the (R²) instead of (r) removes the negative numbers and groups the positive and negative correlation together, as these both can have a negative impact.

Remember we said at the start, you need to avoid some mistakes when measuring portfolio diversification correlation.

If we pick two assets that have the same quote currency, the USD for instance, you would expect two XXXUSD pairs to be correlated but…

WARNING

They will not always be correlated due to differences in the base currency. This can give an unreliable (R²) figure. You can see on the chart in the video there are many outliers. Although we have a nice high (R²) score, you need to be careful.

If we now do the same for two stock index pairs, we can expect these to be correlated. Stock indexes tend to move together, which they are. So even though these two pairs give us similar results we can trust the stock index value more.

When systematically testing various aspects of your trading strategy you need to try to disprove your theories. Yes, disprove. Only by trying to disprove them will you find truly robust parameters and metrics to use.

Which of these 3 pairs do you think has the strongest correlation?

  1. AUDUSD & NZDUSD
  2. GBPUSD & AUDUSD
  3. AUDJPY & USDCAD

Answers on a postcard or just tag us on Twitter at @darwinexchange might be easier.

Brought to you by Darwinex: UK FCA Regulated Broker, Asset Manager & Trader Exchange where Traders can legally attract Investor Capital and charge Performance Fees.

Risk disclosure:
https://www.darwinex.com/legal/risk-disclaimer


Content Disclaimer: The contents of this video (and all other videos by the presenter) are for educational purposes only, and are not to be construed as financial and/or investment advice.

Benefits of Portfolio Diversification

Measuring Correlation to inform your Portfolio Diversification Strategy

Measuring correlation is as important as portfolio diversification itself. We can’t diversify our portfolio without first measuring correlation between the assets and strategies in our portfolio.

In fact, not measuring correlation will potentially lead to the risk we are exposed to increasing. By having exposure in two correlated positions, we are effectively doubling our exposure to this risk.

So how do we measure correlation?

The method for measuring correlation will differ depending on what technique we are using to diversify our portfolio. Remember the four techniques we covered previously in this series. These were:

  • Across asset classes
  • Within the same asset class
  • Across different timeframes
  • Across trading strategies

Throughout this mini-series, we’re going to look at different methods of measuring correlation.

To kick things off; we are going to start with the Coefficient of correlation (r) or Pearson’s r, as it’s also known. Karl Pearson developed the idea. He also founded the world’s first university statistics department at the University College, London.

Pearson’s r measures the strength of a linear relationship between two variables. It does this by creating a line of best fit between variables and measures the distance of each variable to the line of best fit.

This results in a range of results between -1 and +1. -1 would be a negative correlation, 0 would be no correlation and you guessed it +1 would be a positive correlation.

What do we mean by positive and negative correlation?

An example of a positive correlation would be two separate trades in the same direction, on the same asset. When the asset price rises; both the trades will gain, due to the positive correlation between the two trades.

A negative correlation would be simply the opposite. Two separate trades in different directions on the same asset. When the asset price rises one trade will gain whilst the other loses and vice versa if the asset price falls.

Positive and negative correlation can both negatively affect our portfolio.

So why measure them individually?

We don’t have to. There is another way to measure correlation. The coefficient of determination (R²).

This takes Pearson’s r and squares it. Doing so removes the negative part of the range leaving a range of 0 to 1. 0 would equal uncorrelated and 1 would equal either positively or negatively correlated. As both positive and negative correlation will negatively impact the portfolio this way would seem to make sense.

Brought to you by Darwinex: UK FCA Regulated Broker, Asset Manager & Trader Exchange where Traders can legally attract Investor Capital and charge Performance Fees.

Risk disclosure:
https://www.darwinex.com/legal/risk-disclaimer


Content Disclaimer: The contents of this video (and all other videos by the presenter) are for educational purposes only, and are not to be construed as financial and/or investment advice.