Cook’s Distance

PrepNuggets

LEVEL II

Cook’s distance is a measure of the effect of deleting an observation on the estimated coefficients. It takes into account both the leverage and the residual of the observation. High Cook’s distance indicates that the observation has a large effect on the estimated coefficients when it’s deleted.

If the Cook’s D is higher than 1.0, or 2x√(k/n), the observation is highly likely to be influential. The Cook’s distance is useful for identifying influential data points as it measure the change in regression estimate if the observation is deleted.

See also: Influence analysis, High leverage, Studentized residual, Influential data point, Influence plot