The Comparative R2 and Its Application to Self-referencing Models

The traditional coefficient of determination or R is the proportion of variation explained by a regression model versus the variation explained by the mean. This measure does not discriminate well between alternative self-referencing models such as site index curves. The traditional R compares the variation explained by a model with the variation about the mean dependent variable, a very simple model. A generalized R based on the proportion of the variation explained by the self-referencing model versus the variation explained by another simpler (yet more complicated than the mean) model provides better discrimination between candidate models. We call this generalized R the Comparative Coefficient of Determination or Comparative R. Three growth series or plots from the South Africa Correlated Curve Trend Study are used to illustrate the difference between the traditional R and the generalized R.