But First, Correlation. To understand regression to the mean, you must first understand correlation, which is the strength of the relationship between two variables.

With a correlation that low, there will be some mean regression at work. Think of it this way. Seven door-to-door salesmen go out and sell as many vacuum cleaners as they can on a Sunday.

The other key result is the correlation of the two. Regression statistics will typically include an R-squared value. The closer to 1 this is, the stronger the correlation between the returns of ...

By the way, the combined correlation between average sales price and volume is -0.73, which is pretty strong. Now, after adding these into our predictive model, we are running our regression analysis.

Caiyan Li, Hongzhe Li, VARIABLE SELECTION AND REGRESSION ANALYSIS FOR GRAPH-STRUCTURED COVARIATES WITH AN APPLICATION TO GENOMICS, The Annals of Applied Statistics, Vol. 4, No. 3 (September 2010), pp.