Alts University: What is Correlation?

Whiteboard Video Series: What is Correlation?

Watch this short 4-minute educational video to learn more about correlation and what it means for your client portfolios.



As part of our ongoing University of Alts content library, today we're going to cover Correlation.

As always, we'll start with a definition. The correlation coefficient measures the degree to which two assets move in relation to one another. This can include individual securities like stocks and bonds, broader portfolios like mutual funds and ETFs, as well as asset classes and financial markets.

Correlation is measured on a scale of -1 to +1. A value of -1 would indicate perfect negative correlation, meaning one asset always moves the opposite direction as another. A value of +1 would indicate perfect positive correlation, meaning one asset always moves in the same direction as another. A value of 0 would indicate that two assets are completely uncorrelated and, generally, the movements of one will have no relation to the movements of another.

Next, let's talk about why this is an important metric to take into consideration when building portfolios. One of the biggest goals of diversification is to reduce overall risk while also increasing the potential for improved returns; in other words, creating the most efficient portfolio possible.

A common misconception when it comes to portfolio construction is that diversification can be achieved by simply combining a large number of investments. If they are highly correlated to one anther, however, the risk reduction benefit would be minimal. Instead, what an investor should consider when building a well-diversified portfolio is not the number of investments but, rather, the correlation to one another because the less correlated a group of investments are, the greater the effect will be in reducing the overall risk of the portfolio. This sort of uncorrelated diversification can be achieved by adding different asset classes such as stocks and bonds, diversifying across global regions, and also introducing alternative investments like real assets or managed futures strategies.

Next, we'll take a closer look at how these correlation metrics can be used in practice. The actual calculation for correlation is quite complicated. It involves calculating the covariance and standard deviation for a set of subject investments.

Rather than perform the calculation themselves, most investors rely on popular data providers to do it for them. Oftentimes, these results will be laid out in a correlation matrix such as the one shown here. This type of chart can help an investor to see how each of their investments are correlated to one another and help ensure that they are achieving the level of diversification that they are targeting.

In this example, we can see that domestic and foreign equities are generally highly correlated to one another, with a coefficient of 0.95. However, an investment such as managed futures maintains a low correlation to all other included asset classes, with a correlation coefficient no higher than 0.22.

As we mentioned earlier, by building a well-diversified portfolio an investor may be able to achieve more consistent returns over time at a lower level of overall risk. And one important consideration to building this well-diversified portfolio is the correlation of all investments included within.