Last update on .

What is correlation

Correlation (aka: Correlation Coefficient) measures the linear relationship between two variables. In the context of trading or investing, correlation refers to the linear relationship of the historical price data of two instruments over a given time frame.

Overview

Correlation in the context of the stock market is the tendency for two stocks or ETF's to move in the same direction at the same time, or to move in the opposite direction at the same time.

When dealing with correlation, you are essentially asking the question - In the past, how strong was the tendency for two instruments to either increase in price at the same time or decrease in price at the same time. On the other end of the scale, how strong was the tendency for these two stocks to move in the opposite direction at the same time (ie: inverse correlation).

Correlation values range from -1.00 to +1.00. A value of +1.00 means the two instruments are perfectly correlated as a result of the first stock always moving up at the same time as the second one. A value of -1.00 means the two instruments have a perfectly inverse correlation as a result of the first stock always moving up while the second one moves down. A value of 0 means there is no correlation as a result of the two stocks having no statistical relationship in terms of their prior directional movements. The correlation value can range anywhere between those two extremes (-1 to +1) which indicates the strength of the relationship.

Important Points to Remember

There are three important points to keep in mind when dealing with correlation, namely:

  1. Differing Time Frames

    The exact correlation between two stocks or ETF's will differ depending on how much data is used in the calculation. For example the correlation calculated over the prior 1 month will differ when compared to the same calculation performed over the prior year.

    It is important to look at multiple correlation timeframes to put the current relationship into context. Also, keep your investment time horizon in mind. If you are a long term investor, then a 1 month correlation is likely less important than a 1 year correlation. Active traders however would focus more on shorter timeframes while using longer timeframe correlations for context.
  2. Correlation focuses on direction, not magnitude

    Correlation does not tell you how much each stock will move up or down. If stock ABC hypothetically has a +0.85 correlation to stock XYZ, we know that means they have a very high tendency to move in the same direction at the same time. However we cannot say that if stock ABC moves up $1, stock XYZ will move up $0.85. We simply know that if the first stock moves up, there is a very high likely hood the second one will as well.
  3. Correlation is not Causation

    Even with a very strong positive or negative correlation, we must remember that one stock going up does not cause the other to go up. There are often fundamental, company specific and broad market influences that result in correlations being what they are. However ultimately the price of every stock is a representation of the value currently given to that company by all the participants in the market.

    With that said, correlation is an exceptionally powerful tool for portfolio construction, market analysis, diversification, risk reduction and strategic allocation. The key is to understand the meaning behind the numbers and to use that information to aid in constructing a profitable portfolio.

Where to start

When first adding correlation as a factor in your decision process, start looking for logical relationships:

  1. Industry specific fundamental influences. For example oil companies.
  2. Look at Sector or Industry ETF's and their correlation to the stocks that make up those ETF's.
  3. Companies that provide key products and services to each other.

Keep in mind that finding highly positive correlations is not the only goal. In fact there are strong reasons to rather construct portfolio's comprised mostly of low or negatively correlated stocks. This will be the topic of future posts.

To finish off this article, here are 3 charts illustrating highly positive, highly negative, and near zero correlation stocks:

  • Highly positive correlation: VISA (V) - Mastercard (MA)
    Notice how the pair tends to move in the same direction around the same time.

  • Highly negative correlation: Direxion Daily Gold Miners Bear 3X ETF (DUST) - AngloGold Ashanti Limited (AU)
    Notice how the pair tends to move in opposite directions around the same time.

  • Near zero correlation: Portola Pharmaceuticals, Inc. (PTLA) - Sunshine Heart, Inc. (SSH)
    Notice how the pair tends to have no discernible relationship in direction.

For more articles on strategy, tips, education and stock market analysis - click here to subscribe to our free newsletter

  • Stay informed by following us:

Comments

Comments are closed.