Influence Plots
These values provide measures of the influence, potential or actual, of individual runs. The graphical plots provide a better perspective on whether a case (or two) “sticks out” from the others.
Cook’s distance: A measure of how much the entire regression function changes when the ith point is not included for fitting the model. It is essentially the sum of differences in predictions at every point caused by leaving a point out for fitting the model.
Leverage: A measure of how much each point influences the model fit. If a point has a leverage of 1.0, then the model exactly fits the observation at that point. That point controls the model.
Leverage Limits: A run with leverage greater than 2 times the average is generally regarded as having high leverage. Such runs have few other runs near them in the factor space.
The average leverage is the number of terms in the model divided by the number of runs in the design.
DFFITS: A measure of how much the prediction changes at the ith point when the ith point is not included for fitting the model.
DFBETAS: A measure of how much a coefficient estimate changes when the ith point is not used to fit the model. There are separate DFBETA plots for each term in the model.
Report: A table of the numbers used to generate the Diagnostic and Influence plots.