Last update on .

The minimum correlation algorithm is a relatively new asset allocation strategy. The strategy is robust in its design, applies solid diversification principles and performs well in relation to other asset allocation strategies.

Introduction

Considering the fact that diversification has long been one of the most widely accepted principles in finance, it makes intuitive sense to use an asset allocation strategy that incorporates correlation based diversification concepts.

As we know from prior articles such as this one, volatility is reduced when we create a portfolio which comprises of low correlation stocks. A well diversified portfolio requires more than simply holding a large number of assets, it needs to take into account the relationships between those assets, the relative risk contribution of those assets and the possibility of large adverse moves in individual positions.

Risk Reduction

When starting with the low number of stocks, each new stock added to the portfolio provides a reasonable reduction to the overall risk. However as the number of stocks grows, risk reduction gets less and less as a consequence of adding additional stocks to the portfolio. At the same time, correlation starts to have a greater effect on risk reduction.

Said another way, as the number of positions increase, it is no longer the act of adding new stocks which matters but rather the relationship (ie: correlation) between them.

Risk Dispersion

Another key component of diversification is minimizing concentration risk through risk dispersion. What this means is that the goal is to spread risk evenly through the portfolio. Two examples of concentration risk are:

  • You have a $100,000 portfolio consisting of 10 positions. 9 positions are equally weighted at $8,000 each. 1 position is weighted at $28,000.
  • You have a $100,000 portfolio consisting of 10 positions. Each position is equally weighted at $10,000. 7 of the positions have a low correlation to each other. 3 of the positions have a high correlation to each other.

Concentration risk can result not only from incorrect position sizing but also from poor diversification. The goal therefore is to balance risk evenly across the portfolio. This is what is often referred to as risk dispersion.

Minimum Correlation Algorithm

The minimum correlation algorithm is a portfolio allocation strategy that incorporates the above concepts by creating a portfolio where assets are proportionally weighted by their average correlation.

Initially, a correlation matrix is created. For example:

Next, the correlation matrix is converted to a relative scale to account for possible negative values (ie: negative correlations). This ensures the correlations can be compared using the same, positive scale.

Next, positions that are "diversifiers" are weighted more favorably. This is done using a rank-weighted function.

Lastly, the volatility of each asset is normalized to ensure equivalent risk.

Using the MCA and rebalancing when required results in a well diversified portfolio that has been shown to have shallow drawdowns, spreads risk evenly through the portfolio and performs very well relative to other portfolio allocation strategies.

More Information

This article provided a brief overview of the minimum correlation algorithm. For more information you can read the complete white paper here.

Free Trial

The asset allocation tool provided by Diversify Portfolio provides you with an implementation of the MCA to use with your stock portfolio. More information on the tool can be found here.

If you aren't already a member, you can sign up for a free trial here.

Stay Informed

Stay informed by joining our mailing list where you'll receive strategy, diversification and educational articles to help maximize your profit in the stock market while minimizing your risk.

Comments

Comments are closed.