Random Factor Analysis

The bane of statisticians’ existence: outliers.

Outliers are those data points that are...rebelling. All of the other data points might be in a nice cluster, or a fairly obvious line or pattern. Meanwhile, outliers are little dots going against the norm. If statisticians are trying to find a pattern, and they do...except for the outliers...you can see their need to explain those outliers.

This is what random factor analysis is for: trying to determine why random data, or outliers, are acting differently than all the other data points.

Outliers could be a result of measuring error, they might be random anomalies, or they might be due to underlying patterns that the statistician just hasn’t found yet. Random factor analysis helps them try to figure this out.

In business, random factor analysis is a way to solve problems and see things more clearly. For instance, a seemingly random spike in gas prices might not be random at all, but because of some geopolitical event going on with OPEC (the oil cartel, if you didn’t know).