What Is Cluster Analysis?
Cluster analysis is a technique used to group sets of objects that share similar characteristics. It is common in statistics. Investors will use cluster analysis to develop a cluster trading approach that helps them build a diversified portfolio. Stocks that exhibit high correlations in returns fall into one basket, those slightly less correlated in another, and so on, until each stock is placed into a category.
If done correctly, the different clusters will exhibit minimal correlation from one another. This way investors gain all the virtues of diversification: reduced downside losses, capital preservation, and the ability to make riskier trades without adding to the total risk. Diversification remains one of the central tenants of investing and cluster analysis is just one channel to achieving it.
- Cluster analysis helps investors develop a cluster trading approach that builds a diversified portfolio of investments.
- Cluster analysis enables investors to buy and cluster assets with related returns that fit different market segments.
- One of the benefits of cluster analysis is to help protect the investor's portfolio against systemic risks that could make the portfolio vulnerable to losses.
- One criticism of cluster analysis is that clusters with a high correlation in returns sometimes share similar risk factors, meaning that weak performance in one cluster could translate to weak performance in another.
Understanding Cluster Analysis
Cluster analysis enables investors to eliminate overlap in their portfolio by identifying securities with related returns. For example, a portfolio of only technology stocks may seem safe and diversified on the surface, but when an event like the Dotcom Bubble strikes, the entire portfolio is vulnerable to significant losses. Buying and clustering assets that fit different market segments is crucial to increase diversification and protect against such systemic risks.
Stock Selection and Trading Based on Cluster Analysis
The technique can also uncover certain categories of stocks like cyclical and growth stocks. These specific strategies fall under the smart beta or factor investing umbrella. They attempt to capture better risk-adjusted returns from specific risk premiums like minimum volatility, growth, and momentum.
In some way, smart beta or factor investing embodies the concepts of grouping and categorization preached by cluster analysis. The logic of clustering on a single common behavior mirrors the basic methodology behind factor investing, which identifies stocks susceptible to similar systemic risks and share similar characteristics.
It's not always the case that assets in a cluster live in the same industry. Oftentimes, clusters hold stocks from multiple industries like technology and financials.
Criticism of Cluster Analysis
An obvious drawback to cluster analysis is the level of overlap between clusters. Clusters close in distance, meaning a high correlation in returns, often share some similar risk factors. Thus, a down day in one cluster could translate to an equally weak performance in another cluster. For this reason, investors should find and cluster stocks with a large distance between them. That way, the clusters are impacted by different market factors.
That said, broad market pullbacks like the 2008 Recession will throttle the entire portfolio regardless of its construction. Even the most diversified clusters would have trouble withstanding recessionary headwinds. Here, the best clustering can do is minimize the extreme downside losses.